You can't pluck that particular graphic (which
happens to spell "smile" in English) and use it for your own restaurant, or other purposes, because that particular graphic is trademarked by McDonalds.
No, copyright law offers protection from having that exact image reproduced, and copyright doesn't have to be explicitly registered with the trademark office and offers a much more limited scope of protection. If you look at the cup, it has the word smile, printed in yellow in a boring sans-serif font, standing by itself with a trademark symbol next to it.
I think that this is way too generic to issue a trademark for, especially since it isn't even the product name. "Enjoy Coke-a-Cola (tm)" is fine, "Enjoy", by itself, shouldn't be. Using a common word like "Windows (tm)" as a trademarked name for a product in a specific market is ok, if a bit questionable given prior functional use of the term in the market, but, to make up an example similar to the McDonald's one, they shouldn't be able to just put the words "easy to use" on the box in some generic font and trademark that.
The current state of Napster is such that
most files share an MD5 sum. An MD5 is calculated using the entire file, and is COMPLETELY different if even one bit is changed.
If Napster started filtering by MD5, it would be a trivial task, on the order of a few hours, to write a daemon or Windows service that monitors a directory, and every 30 seconds changes the id3 comments field on every mp3 to a new, random value. There are 30 bytes in that field, so there would now be something on the order of 1.77e+72 possible MD5s (minus hash collisions) for each individual rip, without interfering at all with the text search capability, playability or identifiability (otherwise) of the files. Also, this number is within a couple of orders of magnitude of the total number of possible MD5 hashes (which are 32 bytes long), so there would inevitably be hash collisions with many other, perfectly legal files even if you SOMEHOW could block all those possible MD5s. (Hint: It's completely ridiculous. There are not enough particles in the visible universe to represent that many permutations even at one particle per bit!)
So when Napster says it's technically infeasable to filter by MD5, they're not lying. As soon as they legally committed to such an attempt, it would become instantly unmanageable if even a small percentage of users deployed the type of tool described above.
While the church/state lines are very sketchy in Utah, the interrelation of the Mormon church and many business in the state is far far worse. Everybody lists their Mormon affiliation on their resume and it's not uncommon for employers to check church references, etc. for a prospective applicant, and not to hire non-Mormons, de facto if not de jure. This is not to mention the enormous "good old boys" type network that means many jobs don't get listed at all, just filled by church buddies and their families.
Even if I got a great job offer from a non-Mormon, international company in Utah, I wouldn't take it, since as an atheist, if I ever got laid off or wanted to change jobs, I would have maybe 10% of the options available to devout LDS chuchgoers.
And as an atheist, I'm still legally protected by federal anti-discrimination laws, so the worst abuses are prevented or hidden. Banish the thought of ever living and working there if you're openly homosexual. I'd also think twice if I were a minority. It's disgracefully recently that blacks were even allowed to become LDS ministers, and the church doctrine is fundamentally racist. (e.g. Native Americans' darker skin and pretty much their entire culture being a punishment from God is part of the basic story of the Book of Mormon)
Yes, you can make $1 per envelope stuffed. Just take out a classified ad that says "Make money stuffing envelopes at home! Send $1 for information to P.O. Box 1234" Collect the dollar from every sucker you hook, and send them a copy of this set of instructions. It's not even fraud!;)
What you say is true, but in my experience working on a relatively large Java project with over a dozen developers with very little C and Java experience, we give almost zero attention to memory management issues. We have a large servlet application that we run hundreds of thousands of transactions through with no appreciable growth in memory usage, so it's not like we just ignore problems; I can confidently say there aren't any of any consequence.
Pretty much the only place we have to care about resource and reference cleanup is with JDBC. Those instances when we have had problems (always with JDBC ResultSets not being closed), a $500 tool called OptimizeIt has been able to show us what line of code caused the problem with less than five minutes of total time devoted to the problem.
So, while Java is not perfect, in my experience it cuts down by 99% the total amount of time and effort you have to devote to memory related issues. With a dozen Java newbie developers adding 60,000 lines of code to an application over a year period, as the lead programmer who deals with such problems, I've spent two hours on memory management issues. No memory checking tool for C and C++ do that.
The best games were the classic ones. They had good gameplay, not advertising driven spiffy graphics designed to hook you with five minutes of playing in Kmart but leave you bored fifteen hours later. (so you'll buy another)
My favorites were:
nethack, hack, rogue played over a 1200 baud modem link to the local community access SCO system
Ultima V
the DOS version of Spacewar on a Victor 9000
Old school Apple ][ educational games like Rocky's Boots and Snooper Troopers.
There have been no new orders for nuclear power plants in the US since 1977.
The reason is not primarily environmental protests, but economics. Nuclear power plants have not been profitable to operate. The extreme complexity of the systems makes them difficult and expensive to maintain, and they are often offline as a result. Also, the cost of constantly upgrading safety systems to the latest standards was economically ruinous.
Many existing nuclear plants are being shut down before their life span is out. In fact, one of the main reasons why electrical companies are pushing so hard for deregulation is that they're losing big time on their nuclear plants but aren't allowed to pass those costs on to consumers or shut down the plants, they have to keep operating them at a loss.
Also, the federal government has been, for decades, refusing to honor their promise made to the nuclear power industry to provide a national long term nuclear waste storage facility, so most plants are keeping all their wastes on-site. This is another expense and most plants are running out of space. They can't operate without somewhere to put spent fuel.
Effectively, there isn't a US nuclear power industry anymore, and nobody in the power generation business wants one.
I'm not sure what you mean by the following, as I'm not a.NET expert...
With.NET objects created in C++ can inherit from objects created in any other language including Java. In.NET, local (none networked) cross-langauge object reuse is easily done unlike in broken-ass CORBA.
but I think that trying to do black-box implementation inheritance between cross-language components, if it is possible at all, would be a horrible hack and a BAD idea. The experience of most OO practicioners has been that inheritance (even in single language systems) is often overused and that, especially for distributed and component based systems (where inheritance creates deployment nightmares) aggregation and programming to interfaces are to be preferred to inheritance for building reusable and maintainable systems. Read Clemens Szyperski's book on component software if you prefer an expert opinion. (He's a Microsoft Research Architect.)
I will agree with you that the support services for CORBA are usually vendor specific and rarely there at all. EJB containers remedey most of this while being cross platform at the same time and available from multiple vendors. Also, there are already products coming to market from vendors like IONA for providing EJB like service containers for objects written in C++.
.NET may provide excellent services for many languages, but you can bet they're not going to be provided on any non-Microsoft platform or from another vendor. Look at DCOM; it supposedly is available on UNIX, but it's a minimal subset of core functionality with none of the support services you'd need to build a real application.
Finally, while not all programming languages are created equal, I think most would agree that you're far more likely to be screwed by being locked into a single platform or with a single vendor than by being stuck with a single language.
Java shackles developers by forcing them to use the Java[tm] platform for all development in all three tiers of a client-server application if they plan to use the Java[tm] language for any aspect development
Not true. CORBA has bindings right now for just about as many languages as.NET is planning to support, and these systems can all interoperate. In fact, Java's network and component specifications are going towards a more language neutral format with RMI over IIOP and the next generation CORBA specs and products that allow IIOP access to EJBs and deployment of EJB-like services in any language.
I know this is true because I write Java applications in a three tier system that use C++ components in the middle tier and PL/SQL code in the database tier. We also have Perl code that calls Java components in the middle tier.
There are also many languages that can be compiled into Java bytecode and use Java classes and services.
The real facts are that Java probably gives you more choices and makes it easier to use systems written in other languages and on other platforms than any other language. (C may be slightly more ubiquitous, but much more difficult)
AltaVista, Google and Alexa all have archives of basically the entire web. Alexa, at least, explicitly views this as a function of their company, and boasts of having the largest "library" in the history of humanity.
Many major university computer science departments also have whole-Web archives for the purpose of running siumlations of spiders and other automated information collecting and processing tools.
The main problem is that this information is not always publicly accessible and is within the long arm of the lawyers. Maybe the best way to implement this would be to arrange to have somebody like HavenCo purchase these snapshots on a monthly basis, keep them in near-line storage and move censored content that is deemed important by the Information League back "into print".
I have a social life, and I would still appreciate having a more entertaining and person-friendly work environment.
For example, I live more than an hour's commute from my workplace. I often have social activities near work in the evening. It doesn't make sense for me to go home in the interval, and I'd love to have a couch to nap on or a video game or pool table to play with until my dance lesson or date begins. I might also get some more work done if I could relax for a bit and get refreshed instead of burning out and just working through like a zombie until normal quittin' time.
I also consider the people I work with to be my friends and think tha you should be able to work in an environment where you LIKE going to work. Not that you should do it to the exclusion of family or other things, but liking your job and considering your coworkers your friends doesn't make you a loser. Work is a part of my life, not something I do so I can have a life. Making it more pleasant and fufulling isn't a ploy to steal from the other things I do, it's making that eight hours a day away from my family, friends and hobbies worthwhile in and of itself.
Any programmer worth his salt has heard of, if not read, Donald Knuth's "The Art of Computer Programming". All the examples are written to a hypothetical target machine, MIX, in a generic assembly language. You can get simulators for it for many platforms.
Of course, this generic computer is of 1970's vintage capacity, and wouldn't be much use to many programmers today except as an academic exercise, but he's working on a new version, MMIX, that more closely resembles a 64-bit word RISC processor.
The thing is, on modern processors, it is practically impossible for anybody but an autistic savant to really write efficient code at a machine level. With multiple execution units, long piplines, VLIW processors like the Crusoe, out of order execution, etc., the very idea of trying to control the hardware directly for efficiency's sake is insane. You'll never be able to do it without an enormous intellectual investment in understanding how each chip works, and then your carefully hand-optimized routines would be wasted effort when the next generation chip comes out (probably about two weeks after you've finally figured out the last one).
This is a perfect example of where the hacker ethos of not reinventing the wheel comes in. Let the compiler designers invest that intellectual effort and just use their work.
Gore isn't running primarily on a personal integrity platform. Bush knows that more Americans agree with Gore on the issues and think that Gore is more intelligent and more competent for the job. Therefore Bush has based much of his campaigning and message on his personal integrity and honesty, which, since it seems to be so bad, deserves special attention in proportion to the amount he himself touts it.
Having played around only a little bit with my Mindstorms kit, I can say that it's quite easy to be surprised and fascinated by the behavior that "emerges" from even simple programs utilizing biological type feedback loops.
My only gripe would be the limited storage capacity and input/output channels, but since I've come nowhere near exhausting those on a standard RCX yet, I can't really complain too loudly. Still, I find myself feeling like Turing in the early days, musing about just what would be possible if I had a WHOLE MEG to play with.
So, has anybody published a hack for adding more memory to your RCX? How much can the Motorola chip address?
Sun also provides "end to end" solutions for Web apps. Compare Solaris + iPlanet + Servlets to the Microsoft solution.
iPlanet administrative server must run on a different port from the user server. There is almost no access to Web app level configuration from this menu. (just servlet properties, which you'd have to restart the server to take effect, which requires a password)
iPlanet runs as an app in user space. When installing iPlanet, it warns you that the server should run under an id that has extremely limited permissions at the OS level. "nobody:nobody" is the default setting for this userid.
Because of this partition between Solaris and the Web server, it is nearly impossible for code attacking the webserver to root the box. Even getting a shell as nobody is not too useful.
On the web app side, servlets run in a security sandbox that can be custom tailored to limit access to outside resources. The default settings in iPlanet do not allow file or OS level access from servlets. In fact, the setting to turn this on isn't even in the default config file or admin interface. You have to look it up, know what it is and how and where to add the parameter by hand.
Automatic memory management and array bounds checking in Java prevent the most common form of attacks from being effective. (the app may crash, but it won't compromise your server)
There is still room (there's always room) for poor configuration and insecure apps to cause havoc, but in comparision to the Microsoft toolset, there is much more attention paid to security, segregation of control, and default settings that put security above ease of use.
While the average end user may prefer the ease of use to security, critical civilian sites like NASDAQ and other financial institutions just shouldn't be using products with that philosophy.
To market and sell these products to these types of end users (even a company as huge as MS knows when somebody like NASDAQ is using their software) is irresponsible. To allow an application configuration like that is even more irresponsible. (you can bet that NASDAQ had MSCE's or an MSCSP build this, not somebody's 16 year old nephew)
Sun, in contrast, sends auditors/admins to important customer sites like eBay to make sure they're using the software correctly.
I agree that the folks who built this must shoulder a lot of responsibility, but I cannot absolve Microsoft of culpability. Security is an afterthought in their products, rather than a fundamental design principle, and it shows.
This is not just a problem with one little exploit, it is with Microsoft's whole web app model.
Why doesn't anybody realize that for a Web application, the following things shouldn't be the case:
1) Database passwords, admin passwords, ANY passwords shouldn't be stored on the Web server in plaintext.
2) If an application management interface exists at all on the Web server (which I have some problems with), it should always run on a different port than the application itself and that port should be firewalled such that it can only be accessed from trusted (internal) IPs. The content directory structures for the application and application management should also be segregated.
An architecture that stores permissions and passwords and allows access to change these things and modify the application through the same channels that the application is provided is INHERENTLY INSECURE BY DESIGN.
Sorry if I'm ranting here, but as a professional developer working on a financial site this really tweaks my sense of professional ethics. Who designed this crap? Who audited it and said it was OK? Why do people think that Microsoft's architecture aimed at Joe Idiot who wants to put up a web page about his schnauzer fan club without having to learn anything is suitable for use by NASDAQ for cripessake!?!?
Actually, the original Luddites weren't quite so philosophically minded, and weren't really being put of their job by machines so much as by a change of fashion. They were mostly stocking makers, and the stockings they made were of far far superior quality to what machines at the time could produce. They started losing their jobs not because machines were out competing them, but because stockings went out of fashion in America. Once you weren't showing off your stockings as a fashion statement, it made sense to buy the cheaper, but much poorer quality, machine made stockings.
I think a lot of neo-Luddites are rather the same. They don't have any well thought out objections to technology per-se, they're just people who are losing out to the rapid pace of change in the world, driven by human nature as much as by computers, and they lash out at having to change or having their profit stream threatened.
Look at how hard the MPAA fought against VCRs. The movie industry isn't anti-technology, they're about as high-tech as you can get, innovating constantly throughout this entire century. They just had a good profit scheme going and they'd rather try to keep the status quo they've been winning at than work to sieze the new opportunities present.
I'm sure that this forum will be filled with flames about MS software, which is admittedly insecure, but a lot of it is just ignorance, and even the best Unix systems can be made very insecure this way.
A friend of mine worked on a classified project for a DoD contractor, and I was appalled at his stories. He was set in front of a computer, and his boss called away on business before he could give my buddy a login id. The computer was named "Enterprise". On the bottom of the keyboard was a sticky with the word "Picard" on it. Yes, it was the root password. Similar stickys were to be found on the bottom of nearly every computer in the place.
Worse still, they would download very sensitive data from satellites using rsh to a root account with a.rhosts file! When he pointed out that this was probably the LEAST secure method they could possibly choose, they told him that this scheme was the recommendation of a DoD security consultant.
Their entire idea of security seemed to be putting up a bunch of cold war era posters with eagles playing poker against vodka swilling bears and wolves dressed in arabian garb, warning "Don't tip our hand!"
Admittedly, these weren't machines connected to the outside net, but it would've been trivial for any visitor or janitor to get access to EVERYTHING.
Look closely at the code. You'll see that it only blocks where it matches an ad domain and DOES NOT CONTAIN one of the javascript indicator strings.
This is because most browsers flub if they get blocked trying to load javascript. I allow it through therefore, but it can't annoy too much without any images anyway.
You can create an autoconfig file in JavaScript
and tell your browser to bypass the proxy for
ad sites, thus using your firewall for an ad blocker. The only problem is that you have to load the JavaScript code ad sites want to send you. Browsers degrade a page gracefully if they can't find an image, but they usually fail hard if they can't find a bit of JavaScript code.
Here's an example file. Save it as any filename you like, and set it as the location of your "Automatic Proxy Configuration" in your browser of choice.
function FindProxyForURL(url, host) { //If only a hostname, go directly.
if (isPlainHostName(host)) {
return "DIRECT";
}
I remember reading a security survey done
MONTHS ago about this very same vulnerability. They picked out a couple of dozen e-commerce Web sites and found that a shocking number of them had SQL server running on the *same machine* as IIS, with no firewall of the SQL server ports and the default logins and passwords still enabled. They were able to log in and get credit card numbers, addresses and expiration dates, along with any other personal information collected about customers.
Most of the sites that were in this sorry state were systems put together by MCSE consultants.
Now, I don't have hard evidence to back this up, but I think you'd be pretty unlikely to get that kind of sorry ass configuration from IBM, Oracle or Sun certified consultants using Unix systems. (Linux is another story, but they're not even nearly in the same league as Microsoft when it comes to professional services and turnkey solutions.)
The meatspace metaphor is more like hiring a certified contractor from the world's biggest burglar alarm company to install a home security system, and he leaves the default disable code in the system or installs the master override switch on the outside of your house. The alarm company may not be directly at fault, but there is a strong case for negligence/fraud regarding the "certification" program that is really just a marketing tool.
Anybody in the know care to comment on the threads vs. processes debate in a Solaris vs. Linux context?
How much of this boils down to the following:
Solaris: UltraSPARCs kick ass at SMP cache coherency ==> shared context ("threaded") programming preferred.
Linux: Intel chips suck ass at SMP cache coherency ==> small distinction between the models leads to a preference for private context ("process") programming
You bring up an interesting point. I think you'd certainly be on solid ground to copy software to different computers for your own personal use. You can do the same with books and music.
I think the more interesting ramification here is that software is a more directly "functional" sort of thing than traditionally copyrighted materials. It's easy to argue that you don't get more "functionality" out of having two copies of a book you bought. You may get more personal convenience by having a copy for home and for the office, but it still provides you with the same amount of intellectual "power", if you will.
Software, on the other hand, can increase your "power" by copying it. If you never used two computers simultaneously, you'd have a good fair use argument, but what about installing five copies of Windows on a home network where they're all always on? It may still just be for personal use, but there's definitely an qualitative difference between that and ripping a CD, for example.
Also, I wonder how "personal use" extends to a corporation. (A legal "person".) It's probable the whole origin of sofware "licenses" rather than sales is to prevent just this sort of thing by corporate "people".
Slashdot Mirror
No, copyright law offers protection from having that exact image reproduced, and copyright doesn't have to be explicitly registered with the trademark office and offers a much more limited scope of protection. If you look at the cup, it has the word smile, printed in yellow in a boring sans-serif font, standing by itself with a trademark symbol next to it.
I think that this is way too generic to issue a trademark for, especially since it isn't even the product name. "Enjoy Coke-a-Cola (tm)" is fine, "Enjoy", by itself, shouldn't be. Using a common word like "Windows (tm)" as a trademarked name for a product in a specific market is ok, if a bit questionable given prior functional use of the term in the market, but, to make up an example similar to the McDonald's one, they shouldn't be able to just put the words "easy to use" on the box in some generic font and trademark that.
smile(tm)
WTF!? They trademarked the word "smile"? Makes me sick...
If Napster started filtering by MD5, it would be a trivial task, on the order of a few hours, to write a daemon or Windows service that monitors a directory, and every 30 seconds changes the id3 comments field on every mp3 to a new, random value. There are 30 bytes in that field, so there would now be something on the order of 1.77e+72 possible MD5s (minus hash collisions) for each individual rip, without interfering at all with the text search capability, playability or identifiability (otherwise) of the files. Also, this number is within a couple of orders of magnitude of the total number of possible MD5 hashes (which are 32 bytes long), so there would inevitably be hash collisions with many other, perfectly legal files even if you SOMEHOW could block all those possible MD5s. (Hint: It's completely ridiculous. There are not enough particles in the visible universe to represent that many permutations even at one particle per bit!)
So when Napster says it's technically infeasable to filter by MD5, they're not lying. As soon as they legally committed to such an attempt, it would become instantly unmanageable if even a small percentage of users deployed the type of tool described above.
Even if I got a great job offer from a non-Mormon, international company in Utah, I wouldn't take it, since as an atheist, if I ever got laid off or wanted to change jobs, I would have maybe 10% of the options available to devout LDS chuchgoers.
And as an atheist, I'm still legally protected by federal anti-discrimination laws, so the worst abuses are prevented or hidden. Banish the thought of ever living and working there if you're openly homosexual. I'd also think twice if I were a minority. It's disgracefully recently that blacks were even allowed to become LDS ministers, and the church doctrine is fundamentally racist. (e.g. Native Americans' darker skin and pretty much their entire culture being a punishment from God is part of the basic story of the Book of Mormon)
Yes, you can make $1 per envelope stuffed. Just take out a classified ad that says "Make money stuffing envelopes at home! Send $1 for information to P.O. Box 1234" Collect the dollar from every sucker you hook, and send them a copy of this set of instructions. It's not even fraud! ;)
Pretty much the only place we have to care about resource and reference cleanup is with JDBC. Those instances when we have had problems (always with JDBC ResultSets not being closed), a $500 tool called OptimizeIt has been able to show us what line of code caused the problem with less than five minutes of total time devoted to the problem.
So, while Java is not perfect, in my experience it cuts down by 99% the total amount of time and effort you have to devote to memory related issues. With a dozen Java newbie developers adding 60,000 lines of code to an application over a year period, as the lead programmer who deals with such problems, I've spent two hours on memory management issues. No memory checking tool for C and C++ do that.
My favorites were:
nethack, hack, rogue played over a 1200 baud modem link to the local community access SCO system
Ultima V
the DOS version of Spacewar on a Victor 9000
Old school Apple ][ educational games like Rocky's Boots and Snooper Troopers.
Apple Computer was sued by Apple Records over exactly this same sort of bullshit.
The reason is not primarily environmental protests, but economics. Nuclear power plants have not been profitable to operate. The extreme complexity of the systems makes them difficult and expensive to maintain, and they are often offline as a result. Also, the cost of constantly upgrading safety systems to the latest standards was economically ruinous.
Many existing nuclear plants are being shut down before their life span is out. In fact, one of the main reasons why electrical companies are pushing so hard for deregulation is that they're losing big time on their nuclear plants but aren't allowed to pass those costs on to consumers or shut down the plants, they have to keep operating them at a loss.
Also, the federal government has been, for decades, refusing to honor their promise made to the nuclear power industry to provide a national long term nuclear waste storage facility, so most plants are keeping all their wastes on-site. This is another expense and most plants are running out of space. They can't operate without somewhere to put spent fuel.
Effectively, there isn't a US nuclear power industry anymore, and nobody in the power generation business wants one.
With .NET objects created in C++ can inherit from objects created in any other language including Java. In .NET, local (none networked) cross-langauge object reuse is easily done unlike in broken-ass CORBA.
but I think that trying to do black-box implementation inheritance between cross-language components, if it is possible at all, would be a horrible hack and a BAD idea. The experience of most OO practicioners has been that inheritance (even in single language systems) is often overused and that, especially for distributed and component based systems (where inheritance creates deployment nightmares) aggregation and programming to interfaces are to be preferred to inheritance for building reusable and maintainable systems. Read Clemens Szyperski's book on component software if you prefer an expert opinion. (He's a Microsoft Research Architect.)
I will agree with you that the support services for CORBA are usually vendor specific and rarely there at all. EJB containers remedey most of this while being cross platform at the same time and available from multiple vendors. Also, there are already products coming to market from vendors like IONA for providing EJB like service containers for objects written in C++.
Finally, while not all programming languages are created equal, I think most would agree that you're far more likely to be screwed by being locked into a single platform or with a single vendor than by being stuck with a single language.
Not true. CORBA has bindings right now for just about as many languages as .NET is planning to support, and these systems can all interoperate. In fact, Java's network and component specifications are going towards a more language neutral format with RMI over IIOP and the next generation CORBA specs and products that allow IIOP access to EJBs and deployment of EJB-like services in any language.
I know this is true because I write Java applications in a three tier system that use C++ components in the middle tier and PL/SQL code in the database tier. We also have Perl code that calls Java components in the middle tier.
There are also many languages that can be compiled into Java bytecode and use Java classes and services.
The real facts are that Java probably gives you more choices and makes it easier to use systems written in other languages and on other platforms than any other language. (C may be slightly more ubiquitous, but much more difficult)
Many major university computer science departments also have whole-Web archives for the purpose of running siumlations of spiders and other automated information collecting and processing tools.
The main problem is that this information is not always publicly accessible and is within the long arm of the lawyers. Maybe the best way to implement this would be to arrange to have somebody like HavenCo purchase these snapshots on a monthly basis, keep them in near-line storage and move censored content that is deemed important by the Information League back "into print".
For example, I live more than an hour's commute from my workplace. I often have social activities near work in the evening. It doesn't make sense for me to go home in the interval, and I'd love to have a couch to nap on or a video game or pool table to play with until my dance lesson or date begins. I might also get some more work done if I could relax for a bit and get refreshed instead of burning out and just working through like a zombie until normal quittin' time.
I also consider the people I work with to be my friends and think tha you should be able to work in an environment where you LIKE going to work. Not that you should do it to the exclusion of family or other things, but liking your job and considering your coworkers your friends doesn't make you a loser. Work is a part of my life, not something I do so I can have a life. Making it more pleasant and fufulling isn't a ploy to steal from the other things I do, it's making that eight hours a day away from my family, friends and hobbies worthwhile in and of itself.
Of course, this generic computer is of 1970's vintage capacity, and wouldn't be much use to many programmers today except as an academic exercise, but he's working on a new version, MMIX, that more closely resembles a 64-bit word RISC processor.
The thing is, on modern processors, it is practically impossible for anybody but an autistic savant to really write efficient code at a machine level. With multiple execution units, long piplines, VLIW processors like the Crusoe, out of order execution, etc., the very idea of trying to control the hardware directly for efficiency's sake is insane. You'll never be able to do it without an enormous intellectual investment in understanding how each chip works, and then your carefully hand-optimized routines would be wasted effort when the next generation chip comes out (probably about two weeks after you've finally figured out the last one).
This is a perfect example of where the hacker ethos of not reinventing the wheel comes in. Let the compiler designers invest that intellectual effort and just use their work.
Gore isn't running primarily on a personal integrity platform. Bush knows that more Americans agree with Gore on the issues and think that Gore is more intelligent and more competent for the job. Therefore Bush has based much of his campaigning and message on his personal integrity and honesty, which, since it seems to be so bad, deserves special attention in proportion to the amount he himself touts it.
My only gripe would be the limited storage capacity and input/output channels, but since I've come nowhere near exhausting those on a standard RCX yet, I can't really complain too loudly. Still, I find myself feeling like Turing in the early days, musing about just what would be possible if I had a WHOLE MEG to play with.
So, has anybody published a hack for adding more memory to your RCX? How much can the Motorola chip address?
iPlanet administrative server must run on a different port from the user server. There is almost no access to Web app level configuration from this menu. (just servlet properties, which you'd have to restart the server to take effect, which requires a password)
iPlanet runs as an app in user space. When installing iPlanet, it warns you that the server should run under an id that has extremely limited permissions at the OS level. "nobody:nobody" is the default setting for this userid.
Because of this partition between Solaris and the Web server, it is nearly impossible for code attacking the webserver to root the box. Even getting a shell as nobody is not too useful.
On the web app side, servlets run in a security sandbox that can be custom tailored to limit access to outside resources. The default settings in iPlanet do not allow file or OS level access from servlets. In fact, the setting to turn this on isn't even in the default config file or admin interface. You have to look it up, know what it is and how and where to add the parameter by hand.
Automatic memory management and array bounds checking in Java prevent the most common form of attacks from being effective. (the app may crash, but it won't compromise your server)
There is still room (there's always room) for poor configuration and insecure apps to cause havoc, but in comparision to the Microsoft toolset, there is much more attention paid to security, segregation of control, and default settings that put security above ease of use.
While the average end user may prefer the ease of use to security, critical civilian sites like NASDAQ and other financial institutions just shouldn't be using products with that philosophy. To market and sell these products to these types of end users (even a company as huge as MS knows when somebody like NASDAQ is using their software) is irresponsible. To allow an application configuration like that is even more irresponsible. (you can bet that NASDAQ had MSCE's or an MSCSP build this, not somebody's 16 year old nephew) Sun, in contrast, sends auditors/admins to important customer sites like eBay to make sure they're using the software correctly.
I agree that the folks who built this must shoulder a lot of responsibility, but I cannot absolve Microsoft of culpability. Security is an afterthought in their products, rather than a fundamental design principle, and it shows.
Why doesn't anybody realize that for a Web application, the following things shouldn't be the case:
1) Database passwords, admin passwords, ANY passwords shouldn't be stored on the Web server in plaintext.
2) If an application management interface exists at all on the Web server (which I have some problems with), it should always run on a different port than the application itself and that port should be firewalled such that it can only be accessed from trusted (internal) IPs. The content directory structures for the application and application management should also be segregated.
An architecture that stores permissions and passwords and allows access to change these things and modify the application through the same channels that the application is provided is INHERENTLY INSECURE BY DESIGN.
Sorry if I'm ranting here, but as a professional developer working on a financial site this really tweaks my sense of professional ethics. Who designed this crap? Who audited it and said it was OK? Why do people think that Microsoft's architecture aimed at Joe Idiot who wants to put up a web page about his schnauzer fan club without having to learn anything is suitable for use by NASDAQ for cripessake!?!?
I think a lot of neo-Luddites are rather the same. They don't have any well thought out objections to technology per-se, they're just people who are losing out to the rapid pace of change in the world, driven by human nature as much as by computers, and they lash out at having to change or having their profit stream threatened.
Look at how hard the MPAA fought against VCRs. The movie industry isn't anti-technology, they're about as high-tech as you can get, innovating constantly throughout this entire century. They just had a good profit scheme going and they'd rather try to keep the status quo they've been winning at than work to sieze the new opportunities present.
A friend of mine worked on a classified project for a DoD contractor, and I was appalled at his stories. He was set in front of a computer, and his boss called away on business before he could give my buddy a login id. The computer was named "Enterprise". On the bottom of the keyboard was a sticky with the word "Picard" on it. Yes, it was the root password. Similar stickys were to be found on the bottom of nearly every computer in the place.
Worse still, they would download very sensitive data from satellites using rsh to a root account with a .rhosts file! When he pointed out that this was probably the LEAST secure method they could possibly choose, they told him that this scheme was the recommendation of a DoD security consultant.
Their entire idea of security seemed to be putting up a bunch of cold war era posters with eagles playing poker against vodka swilling bears and wolves dressed in arabian garb, warning "Don't tip our hand!"
Admittedly, these weren't machines connected to the outside net, but it would've been trivial for any visitor or janitor to get access to EVERYTHING.
This is because most browsers flub if they get blocked trying to load javascript. I allow it through therefore, but it can't annoy too much without any images anyway.
Here's an example file. Save it as any filename you like, and set it as the location of your "Automatic Proxy Configuration" in your browser of choice.
function FindProxyForURL(url, host) {
//If only a hostname, go directly.
// Remove a few ads /*Your proxy addr and port here*/";
if (isPlainHostName(host)) {
return "DIRECT";
}
if (
(
url.indexOf("/RealMedia/") > 0
|| url.indexOf("ads.x10.com") > 0
|| url.indexOf("ads3.zdnet.com") > 0
|| url.indexOf("/ads/") > 0
|| url.indexOf("/Ads/") > 0
|| url.indexOf("/adverts/") > 0
|| url.indexOf("/adserver/") > 0
|| ( dnsDomainIs( host,"doubleclick.net") && url.indexOf("/adj/") == -1)
|| dnsDomainIs( host,"focalink.com")
|| dnsDomainIs( host,"adbureau.net")
|| dnsDomainIs( host,"ads.imgis.com")
|| dnsDomainIs( host,"ad.preferences.com")
|| dnsDomainIs( host,"view.avenuea.com")
)
&& url.indexOf(".js") == -1
&& url.indexOf("jx.ads") == -1
&& url.indexOf("js.ng") == -1
&& url.indexOf("jsad") == -1
&& url.indexOf("jscript") == -1
&& url.indexOf("addyn") == -1
&& url.indexOf("type=script") == -1
)
{
return "DIRECT";
}
else
{
return "PROXY
}
}
Most of the sites that were in this sorry state were systems put together by MCSE consultants.
Now, I don't have hard evidence to back this up, but I think you'd be pretty unlikely to get that kind of sorry ass configuration from IBM, Oracle or Sun certified consultants using Unix systems. (Linux is another story, but they're not even nearly in the same league as Microsoft when it comes to professional services and turnkey solutions.)
The meatspace metaphor is more like hiring a certified contractor from the world's biggest burglar alarm company to install a home security system, and he leaves the default disable code in the system or installs the master override switch on the outside of your house. The alarm company may not be directly at fault, but there is a strong case for negligence/fraud regarding the "certification" program that is really just a marketing tool.
How much of this boils down to the following:
Solaris:
UltraSPARCs kick ass at SMP cache coherency ==> shared context ("threaded") programming preferred.
Linux:
Intel chips suck ass at SMP cache coherency ==> small distinction between the models leads to a preference for private context ("process") programming
I think the more interesting ramification here is that software is a more directly "functional" sort of thing than traditionally copyrighted materials. It's easy to argue that you don't get more "functionality" out of having two copies of a book you bought. You may get more personal convenience by having a copy for home and for the office, but it still provides you with the same amount of intellectual "power", if you will.
Software, on the other hand, can increase your "power" by copying it. If you never used two computers simultaneously, you'd have a good fair use argument, but what about installing five copies of Windows on a home network where they're all always on? It may still just be for personal use, but there's definitely an qualitative difference between that and ripping a CD, for example.
Also, I wonder how "personal use" extends to a corporation. (A legal "person".) It's probable the whole origin of sofware "licenses" rather than sales is to prevent just this sort of thing by corporate "people".