There are some cases where dual-booting is more advantageous than virtualization. Virtualization takes a heavy hit on RAM (I tried Parallels on my MacBook with a measly 512MB of RAM, my 1.83GHz dual core computer felt like my old 8MHz Mac SE), and if you're strapped for cash and don't have much memory, it's better to just dual-boot where the OS has full access to all of the RAM needed.
That reminds me to invest in a upgrade to 2GB of RAM soon.
So, we trashh the government schools. What then? Who educates the people who can't afford a private education?
What if schools were privately owned but the tuition of the students were subsidized by the government? You know, kind of like how the federal government gives grants to students from lower-income families to cover college costs, public or private? The poor will still be able to go to school, but the schools are no longer government owned. Supporting the privatization of education is not the same as supporting the withdrawal of government funding for education.
There. That fixes the argument. There is a big difference between society and government. Society is simply a collection of people, whereas government is the ruling force of a jurisdiction of land. In some cases the society and government are somewhat intertwined, whereas in other cases the government is far removed from the society that it is governing.
That's because, unlike the old Bell Labs nor Google, Microsoft doesn't really capitalize from its research. Look at the research with Singularity, for example. As a future computer science researcher (I'm just a sophomore in college now), I would love to get my hands on a system like this. Finally something new that isn't based off of nearly 40 years of Unix. The goals are quite noble and innovative, and I'm glad that Microsoft is doing systems research, something that seems to have been neglected in computer science for some time (Rob Pike talked about that in his "System Research is Irrelevant" talk back in 2000). However, Vista makes use of absolutely none of this technology, and Microsoft doesn't seem to want to incorporate any of this research in Windows at all.
MS can be so much better if they actually applied their research to Windows, its flagship product. But since Microsoft has already accomplished its goal, have 90% worldwide marketshare on operating systems, I guess it can get away with incremental improvements every half decade or so. It's not that Microsoft doesn't innovate, look at the research. That's innovation. It's that Microsoft doesn't want to capitalize its innovations and is content on sticking with its Windows (and Office) monopoly.
Project management is not a computer science skill; it's a software engineering skill (software engineering != computer science, although I feel, by experience, that software engineering skills are very important for computer scientists to learn). However, there are schools now that offer software engineering degrees which do explain some of the methodology and practices behind software engineering. Project management does fit in with a software engineering degree, exactly in the same way as project management techniques fit in with the traditional engineering disciplines.
I believe that for most industry positions in the software industry, a software engineering degree would be a better fit than a computer science degree, since the software engineering degree better prepares students for some of the conditions and practices that they'll see in industry. At my school, the software engineering degree also contains courses on teamwork and other engineering practices (sometimes alongside traditional engineers). A computer science degree, by definition, is geared toward people who want to become computer scientists. Computer scientists are focused more on research than software engineers, who are focused on building software. A computer scientist is to a software engineer as a physicist is to an electrical or mechanical engineer. Scientists are focused on expanding knowledge and doing research, whereas engineers are focused on applying scientific knowledge to practical uses.
Read the article again. The article talks about how IIT graduates are doing well in the industry because of their high quality of teaching. The main focus of the article, however, is on other Indian universities, not IIT (which is one of the best schools in the world).
There is only one problem with the Libertarian Party: much of the official party's platform needs to be toned down from some of the borderline anarchocapitalistic stances that the Party takes on certain issues. I feel that the Libertarian Party can make a big splash as a party that stresses the synergy of protecting civil liberties and fiscal conservatism, rather than a party that wants to turn the USA into Ancapistan overnight. Plenty of fiscal and economic conservatives disgruntled about the direction that the Republicans are taking, and aren't fond of the growth of social conservatism would be very interested in a moderate Libertarian party. Liberals who don't like high taxes and heavy government spending, and are disillusioned with the idea that the government would fix economic issues, would also like the party. (I am personally an ex-liberal who turned into somewhat of a libertarian).
The only worthwhile part of Knuth's work is the first book on sorting, and even then it is dry. Just because he can sling nasty formulas around when comparing algorithms doesn't mean a lot. These days we are using hardware that resemble DSPs and locality of reference problems can outweigh counting the number of multiplies by a long shot. Just learning O notation, what is an L2 cache miss and *to actually profile your code* instead of theorizing about it is much more important. His formulas aren't relevant on modern hardware.
The Art of Computer Programming is about algorithm analysis and development, not about computer architecture (caching) or software development (profiling). TAOCP, combined with Concrete Mathematics, the MIT algorithms book, and a good text on combinatorial algorithms should be more than enough to satisfy the needs of a computer scientist or seasoned software developer who needs all of the good algorithms that he or she can get. TAOCP isn't a book about all of computer science, but it isn't trying to be one either.
The K&R's C Programming Language is only useful to people who already know C. And C is such a small language that you shouldn't need a reference book once you know it. However, if you are a language geek and like that kind of thing, get the actual C standard and read it. Seriously, it isn't that big.
K&R is a useful book for C reference; in fact, I consider it as an explanation to the C standard. I refer to K&R, not necessarily the standard, for reference whenever I am stuck on a certain C concept. (Compiler construction is different; you should follow the standard to the letter). There is a reason why the K&R book hasn't been updated since 1989; other than C99 extensions, the book is a very good explanation of C as is. I learned C largely from this book (although, to be fair, I used an O'Reilly book to learn the basics of C before taking a class that used C. The professor in that class didn't assign a book, so I sought out the best book according to ratings: K&R. My book is worn out out of 2 years of use, but I love it).
Why would you carry your groceries anyway? Are you too damn macho to use a shopping cart.
In California, taking a grocery cart out of the parking lot of a store is illegal. It is called shopping cart theft; you're only supposed to use the shopping cart in the store and within the parking lot. So, using a shopping cart outside of the store is not a legal option. Any other suggestions?
Well, I would argue that automobiles are a symbol of freedom. Automobiles allow us to get from door to door, without having to walk for over 5 minutes to a transit stop and without having to transfer buses a few times if you're going longer than a few miles. Automobiles allow us to travel anywhere we choose, instead of wherever the bus or train line goes. With an automobile, getting a load of groceries is easy. Try getting groceries without a car; you wish that you had one by the time you finally carry your groceries over (or give up and pay the taxi). Automobiles allow us to mind our own business. I don't want to listen to a passenger curse somebody out on a cell phone, or deal with some of the seedier characters on the bus or train (this is actually a common occurrence riding public transportation in Sacramento). Finally, with an automobile, we don't have to revolve our lives around other peoples' schedules. You can leave whenever you feel like it with a car. You can't do that with public transportation.
Am I saying automobiles are perfect? No. They are energy inefficient, for one, and I wish hybrid technology will further improve. Our roads are getting much more congested everyday, and our political leaders (in California) refuse to do anything about it (they have refused to build any roads on a large scale statewide since the Jerry Brown administration started in the mid-1970s. Don't get me started. To make a long story short, the Jerry Brown administration dismantled California's grand statewide freeway plans of the 1950s and 60s [which would have been completed by the 1990s], helped cancel many freeways in metro areas [Sacramento, Los Angeles, Bay Area except San Jose, and other areas], and left current construction projects abandoned. Thanks to his administration, and the lack of will from subsequent administrations to roll back his changes, California's highway system went from the best to the worst. Read more about it here.). However, we, as Americans, appreciate to mobility and convenience that we get from automobiles, and I do not feel that public transportation is as flexible or as convenient as automobiles are. We are willing to suck up congestion and higher gas prices in order to ride our automobiles.
Anyway roads are a very inefficient and wastefull mode of long distance transport. THird world countries would be better off with India's example which has the worlds largest rail network
As a road geek, prove this. Now, I know that roads cannot compete with bullet trains when it comes to speed and delivering passengers. (I'd like to see some more bullet train development in this country; California proposal of a train running from San Francisco to Los Angeles via the Highway 99 corridor is a start). However, I believe that roads, especially in the form of expressways and freeways, are still an important and optimal mean of transportation for distances ranging from 5 miles to 500 miles. Freight trains also cannot compete with trucks for getting goods from manufacturers, farms, and shipping ports to your local store. With freight trains, the goods still must make it from the train station to the store.
Besides, there is nothing like the free feeling that you get from driving a car or even riding along in one. Roads can take you almost anywhere. I love the freedom of travel that you get from riding on the many roads that are available. Riding in a car on a long road trip feels like an adventure; something exciting is about to occur. Riding in an Amtrak train (I've never rode a bullet train, so I can't talk about that experience), in my opinion, just don't give me that same exhilaration that riding in a car gets. It can sometimes be a close to grinding experience, especially when your trains get delayed (I'm an occasional Amtrak rider, so I had these experiences first hand). I also find certain roads to be beautiful achievements of civil engineering (go down to Southern California and see the freeways in Orange County; it's impressive). You get some emotion out of fellow road geeks. Now try to get that same emotion out of train structures.
To put it this way, trains are an important method of transportation, and bullet trains are unmatched in their speed and efficiency. But do not neglect the importance and utility of well designed roads. Roads are unmatched in the ability to be able to get from door-to-door without having to transfer, switch modes of transportation, revolve your life around a posted schedule, and other inconveniences.
The CS people really don't care about software dev cycles and databases, (emphasis mine)
Databases are a very important part of computer science. Databases are fundamental tools for many computer science (and other) research projects, and databases themselves are still being researched.
I still believe that a computer science and software engineering split is a great idea. In fact, at my school, the computer science department has two separate programs: computer science and software engineering. (A computer engineering degree is also offered, but it is its own department). Computer science is for people who want a more general exploration of computer science topics, software engineering is for people who want to know about the software engineering process, and computer engineering is a combination of computer science and electrical engineering.
Linux may win a war versus Microsoft on the server, but that is not where Microsoft is most powerful. Microsoft is still the 800 pound king gorilla of the desktop, and Linux still has a way to go before it unseats Microsoft. Heck, OS X has been out for over five years. Many users who have used both OS X and Windows claim that OS X is superior, and many have switched. However, OS X has barely pinched Microsoft, and Microsoft still enjoys 93% or so marketshare on the desktop (5% goes to Mac, and the rest of that is Linux, BSD, and other OSes).
The reason why Windows still hasn't been unseated is that too many people have software that is Windows-only. Businesses still rely on in-house Windows programs that were created by some programmer many years ago who is long gone, and cannot afford porting it to another operating system. Sometimes you'll see a business run an old Windows 3.1 or even DOS application since there is no replacement, and since it is good enough for them to not worry about porting it or creating a clone of it. Engineers aren't dropping AutoCAD anytime soon, and AutoCAD is Windows only. Engineers, being well known for their pragmatism, stick to Windows. Graphics artists on both the PC and the Macintosh who rely on Photoshop, Quark Express, and Dreamweaver are not going to move to Ubuntu and use the GIMP, Scribus, and nvu (yes, those open source products are good, but their commercial competitors are very good and are worth the $$$ that you pay for them; they end up saving you $$$ with their features and ease of use). And developers who want some food on their table better know something about Win32,.NET, and other Windows technologies. In most non-CS fields, you cannot avoid Windows in the professional world, and Windows has became a fact of life in many careers.
So, what is the open source community going to do about this? A great operating system with all of the bells and whistles isn't enough for most people. Once again, OS X is considered the best operating system by many people, but some applications haven't been ported yet (or won't be ported), which doesn't leave OS X as an option for those people. The open source community needs to start polishing up their offerings and get started on some new stuff (an AutoCAD replacement will get engineers off of Windows, for example). GIMP can use some improvement. OpenOffice should be more modular and faster. Dia needs to start looking like OmniGraffle or Visio. There needs to be some sort of OSS equivalent to Visual Basic (what I mean by that is ease of developing GUI applications). I recommend the same with other Linux applications.
Remember, the key to operating system adoption is applications. Look at MS-DOS, for example (back in its heyday). It was hard to use (compared to the Apple Macintosh at the time), very rudimentary (compared to other OSes in the 80s like Unix, NeXTSTEP, and VMS), and can only run one application at a time. But it ran Lotus 1-2-3 and WordPerfect, and it ran all of the other applications that business users wanted. Home users wanted the computers that business users had, so they got that too. Ditto for Windows 3.1. NeXTSTEP knocks the socks off of Windows 3.1. But who had the applications?
Your OS can be the easiest to use OS in the world. It can have microkernels with the best scheduling and load-balancing algorithms that exist. It can utilize all of the systems research published in the ACM and IEEE journals within the past two years. It can be so secure that it would be the envy of Homeland Security and would make Symantec and McAffe angry (they can't sell protection for it). It can even have a mass advertising campaign with beautiful angelic models praising the product. But if it cannot run the applications that they want, then it is just a waste of hard drive space and time as far as they are concerned.
I still like the idea of a CS bachelors, but in the end maybe you are right. Computer Science should be a graduate degree, after a bachelors in Software/Computer Engineering.
I don't like that idea, because prospective computer scientists (not software engineers or computer engineers) would not have an undergraduate major reflecting their future career goals. The more theoretical of them (those going into algorithm development rather than systems) would be forced to become math majors, as that is the closest thing to theoretical computer science (and while math is a fun and great subject to learn, mathematics and computer science are different, although related). However, I believe that at some schools the computer science degree has tried a "one size fits all" approach, with mixed results. I believe that there should be four separate computing-related majors:
Computer science: for the future computer scientists of the world. More theory-bent than the rest of the degrees, but has enough practical applications to understand the practicality of the field
Software engineering: for the future software engineers and developers of the world. Very practical and features software development methodologies and practices.
Computer engineering: for people who want to get their hands dirty with hardware, and to learn how hardware and software interact. Kind of a cross between electrical engineering and computer science. Also a very practical major.
Management information systems: for people who want a mixture of computer programming and business skills.
I prefer this system to getting rid of the general computer science degree. Keeping a general computer science degree for the future computer scientists of the world, while creating degree programs for those who want to specialize in software engineering, computer engineering, or business, better supports the diversity of career goals and academic goals of students interested in some computer-related field.
But that is not a computer science degree. Now, I believe that a computer science degree should show practical applications to real-world situations. However, theory should not be neglected. Courses such as algorithms, theory of computation, and computer architecture (which I noticed was omitted from your list; it can qualify as hardware, but it is important to know computer architecture, especially when doing systems programming) are very important for both software developers and computer scientists. GUI programming and client/server programming are not necessary in a computer science curriculum; GUI programming can possibly tie in to a HCI course, however.
Being taught all theory can be bad because you wouldn't know how the theories tie in to practical applications. But being taught all practice is just as bad, because you wouldn't know how all of the applications truly work. A good computer science program should have a balance of theory and practice.
Exactly, and I'm speaking as a libertarian. The problem with the Libertarian Party is that the official platform sounds like something that comes out of an anarchocapitalist book. Radical libertarianism and anarchocapitalism seems to cause more harm than good. I am not a supporter of the welfare state, but I do support some level of state-level assistance for the poorest of people. I happen to like anti-trust regulations (they keep the market free from the control of monopolies). I also like the Interstate Highway System (and state highway systems), and I would like to see more work done in my state (California) with freeway construction, the proposed bullet train, and other infrastructure improvements. I happen to agree with anti-discrimination laws and other civil rights laws. I believe in state-funded public education, with school choice and eventually vouchers. Those are some positions on issues where the Libertarian Party platform differs from my personal beliefs.
Personally, I wish the federal government would return to its Constitutional roots of limited government, fiscal responsibility, and civil liberties. The federal government has amassed far too much power in the past century. The Income Tax Amendment of 1913 and the Federal Reserve gave the federal government the income collecting mechanism needed to expand the government, and the government exploded in size during the FDR administration. The federal government hasn't been the same since. The original purpose of the United States of America was to focus on the states, not America. The power should return to local communities and states, not the federal government. This country wasn't called the United States of America for no reason.
On a state level, I would like to see more pragmatism instead of ideology. Speaking in terms of California politics, California needs to return to the days of the Pat Brown administration of the 1950s and 60s. Back then, California built the most impressive highway system in the world at the time, organized the higher education system (which features the top-notch UC system, the ubiquitous CSU system that has educated hundreds of thousands of Californians, and the community college system for everybody who wants a higher education), built the aqueducts, and did that all while still remaining quite fiscally prudent. California also had the best public school system in the country at the time. Nearly 50 years later, we Californians live in a shadow of those days. Our highways are still largely of 1960s vintage and are crumbing and in need of improvement, our great highway plans of the 50s are incomplete (freeway plans have been largely abandoned since the beginning of the Jerry Brown administration; don't get me started on him), our public schools are among the worst in the nation, and businesses are leaving California thanks to their sky-high taxes and the high cost of living due to lack of affordable housing (environmentalists stop developers from building on open space near the metro areas, so developers respond to them by building houses 30 miles up or down the road, resulting in crazy traffic congestion).
On a state level, I would like to see their government kind of like the Japanese government. The Japanese government is a model of pragmatism, in my opinion. They are able to fund quite a bit of services (such as highways, bullet trains, and even universal health care) while keeping taxes in check and by not running up huge budget deficits. The government also firmly believes in right-wing economics, such as privatization of certain services (universites have been partially privatized, the postal service has just been privatized, etc.), and they also believe in fiscal conservatism. The Japanese public K-12 school system is the envy of the world, and their universities are very good. On a state level, we need to be more like them.
As for the Libertarians, they need to be much more pragmatic, while at the same time not sacrificing the core of their principles: civil libertarianism fused with fiscal conservatism. A pragmatic version of libertarianism would almost certainly make inroads with much of the population segment that are not extreme leftists nor extreme conservatives.
Anybody can create a virus for OS X, and it can run perfectly. The biggest problem would be how it can be able to spread to other machines.
On Windows, it isn't viruses that plague Windows, but it is worms, spyware, and adware that affects that platform. All it takes to be infected with a computer virus on any platform is to not be vigilant about the data that you download. Being infected by spyware and adware, however, relies on the security of the browser, and being infected with a worm relies on the security of the operating system's Internet connectivity.
OS X remains relatively secure because its browser does not have hooks to the shell (unlike older versions of Internet Explorer, although I've read that Internet Explorer 7 has been decoupled from the shell), and because its Unix core isn't susceptible to worms (Unix has come a long way since the worm of 1988). OS X also has a firewall, although I just learned that it isn't enabled by default (but turning it on is easy; they should change the default in OS X 10.5).
A demo virus for OS X or Linux isn't news. No operating system can block the execution of a virus unless the operating system has a list of trusted applications that it knows are virus-free. An operating system can prevent worms with better security, and spyware can be prevented by using a secure browser, but viruses cannot be blocked from execution.
That is exactly my point. The Linux kernel hasn't changed too greatly in the past 5 years, but the desktop environments for Linux (and BSD) have improved greatly. Linux is harder to judge since there are so many separate components, so I focused on the desktops, since that is the most fair comparison to OS X and Windows.
Five years between operating system releases is a very long time in the computer industry. Look at how OS X improved in five years (from OS X 10.0 to OS X 10.4, with OS X 10.5 coming out sometime in the spring). Look at how KDE and GNOME have improved over 5 years. Look at some other 5-year periods of time in the computing industry. From 1991 to 1996, we went from DOS and Windows 3.0 to Windows 95 and Windows NT, and that is just on the MS side of things.
Five years without any changes other than security fixes and point updates to software is very sad. (Some would say six years if they believe that XP isn't that much of an improvement over 2000, but I'll use XP as a base here).
Am I the only person appalled by these web interfaces, or even web desktops, being referred to as operating systems? It is technically wrong by a large margin. An operating system is the interface between hardware and software that manages the resources of hardware. Web "operating systems" do not manage any hardware.
I find this usage appalling, and I hope that this terminology doesn't spread and dumb down the use of technical terms.
On most Cocoa applications (those that use APIs derived from NEXTSTEP; i.e., most applications that are from the NEXTSTEP days or are applications made 2001 and later) that utilize large toolbars, the toolbars can be fully customized and can be made smaller. The buttons won't be very small like the Office 97 buttons, but they won't be the larger default size, either. Carbon applications (those that use APIs derived from Mac OS 9 and below; i.e., most Mac applications that existed before OS X, such as MS Office and Photoshop) are based off of the older Mac OS 9 and below standard, which used small buttons for toolbars.
There are some Cocoa applications (Safari, Mail.app, TextEdit, Address Book, and a few others) that use small toolbar buttons rather than large ones. You should be fine. As for removing the icons altogether and replacing them with solely text, it depends on the application. Cocoa applications should be easier in this regard to be customized than Carbon ones (but, then again, I don't have any Carbon applications to test this out with, other than iTunes). I hope that your switch to the Mac goes well.
On my MacBook (1.83GHz Core Duo with 512MB RAM), I usually have the following applications open:
Web browser (Safari). I normally browse one site at a time, but sometimes I have 3-4 tabs, especially when Slashdot browsing.
Mail client (Mail.app).
Instant messaging (iChat). Usually two or three windows: one for the buddy list, and one or two for people whom I'm chatting to
iTunes
Depending on the task on hand, I may have a word processor, text editor, some X terminals and X11 applications (via a SSH tunnel to a school account where I do my CSC homework), iCal, or TeXShop open.
I also notice that the amount of windows that I have open is proportional to the speed and performance of the computer that I'm using. For example, when I was using my Duron 950MHz with 384MB RAM and either Windows XP or FreeBSD, I normally only had my web browser (Firefox or Konqueror), GAIM, my email client (Thunderbird or Kontact), and (on Windows) iTunes opened, and sometimes some terminals opened whenever I was doing CSC homework. I would run almost as many windows on that machine as my Mac, but I would be a bit more cautious about performance. On my 266MHz Pentium II laptop with 64MB RAM running FreeBSD, I normally stick to just one application (usually my web browser, Opera, and even that one application can be demanding. (Running some X terminals, gvim, Opera, GAIM, and another GTK app on that machine is an exercise of patience whenever switching between applications, and I couldn't even begin describing the horrors of using OpenOffice on that machine).
Innovations such as Expose and (on KDE and WindowMaker, the desktops that I used on my BSD machines) virtual desktops make using multiple windows much easier. I tried not to have more than 4 or 5 windows opened on my non-Mac machines because there is no easy way of finding all of them (although virtual desktops do help with organizing categories of windows). Because of Expose, I am able to have up to 15 windows on my screen (I've worked with that many windows once) on my Mac and still work very efficiently and be productive without being lost. Bringing virtual desktops to OS X (something that I miss from my BSD experience) and giving virtual desktops the power of Expose will make my Mac even better to use.
Yes. I downloaded the Office 2007 beta back in May and played around with the ribbons interface. The ribbons interface is very good if you are a beginner and you are trying to find options. However, I am far from an Office beginner; I've used Office since the Office 97 days. I know how to customize the toolbar so that way every option that I've needed was accessible from one click from the toolbar instead of multiple clicks from menu options. The ribbons, however, made many things that I was once able to do with one click now take two or even three clicks to do. It takes up a lot of space on the screen, looks ugly on XP and doesn't conform to any pre-Vista Windows standards (but it blends in with Vista very well), and is uncustomizable and can't be removed (the "if you don't like it, tough" approach). It seems to get in the way of my work, and seems to be a much clunkier solution than the old toolbars/menus or even the Inspector approach on OS X.
My biggest problem with ribbons is that it dumb downs Office. Office is a professional tool used to make spreadsheets, documents, presentations, databases, and more. It is inherently complex, kind of like Photoshop, Quark Express, and other complex tools. In the old versions of Office, getting used to the toolbar options and menus may be a bit tough at first, but after finding the options (which is just a few days), you will be productive easily. You can even customize Office to fit your needs and for you to access your most commonly-used features in just one click. Office 2007 forces people to stay beginners forever. Everything requires extra clicks to do compared to Office 97, 2000, XP, and 2003. There is no way to make them one click options, short of memorizing the character shortcuts (which the ribbons do not show, unlike the menus, which show keyboard shortcuts). Office is inherently complex, like Photoshop. Yet if Photoshop's or Quark Express's interface dramatically changed to be beginner-friendly yet professional user-hostile, you'd see a riot in every design company's office.
Perhaps the ribbon interface is a better experience on Vista, where it blends in with the rest of the interface a bit more. Still, I do not like Microsoft's approach to usability. MS's approach tends to dumb down the interface to the point that it is usable for novices and other people who don't have much computer experience, but very difficult to use for people who know how to use a computer. Apple doesn't dumb down anything; it balances creating a usable OS while making it flexible and usable for advanced users at the same time. You can make your interfaces easy to use without alienating experienced users. That's my problem with Office 2007. It makes things easy for beginners at the expense of experienced users.
Slashdot Mirror
There are some cases where dual-booting is more advantageous than virtualization. Virtualization takes a heavy hit on RAM (I tried Parallels on my MacBook with a measly 512MB of RAM, my 1.83GHz dual core computer felt like my old 8MHz Mac SE), and if you're strapped for cash and don't have much memory, it's better to just dual-boot where the OS has full access to all of the RAM needed.
That reminds me to invest in a upgrade to 2GB of RAM soon.
What if schools were privately owned but the tuition of the students were subsidized by the government? You know, kind of like how the federal government gives grants to students from lower-income families to cover college costs, public or private? The poor will still be able to go to school, but the schools are no longer government owned. Supporting the privatization of education is not the same as supporting the withdrawal of government funding for education.
s/society/government/g
There. That fixes the argument. There is a big difference between society and government. Society is simply a collection of people, whereas government is the ruling force of a jurisdiction of land. In some cases the society and government are somewhat intertwined, whereas in other cases the government is far removed from the society that it is governing.
I don't know if I know my DEC history correctly, but what do scanners have to do with DEC?
That's because, unlike the old Bell Labs nor Google, Microsoft doesn't really capitalize from its research. Look at the research with Singularity, for example. As a future computer science researcher (I'm just a sophomore in college now), I would love to get my hands on a system like this. Finally something new that isn't based off of nearly 40 years of Unix. The goals are quite noble and innovative, and I'm glad that Microsoft is doing systems research, something that seems to have been neglected in computer science for some time (Rob Pike talked about that in his "System Research is Irrelevant" talk back in 2000). However, Vista makes use of absolutely none of this technology, and Microsoft doesn't seem to want to incorporate any of this research in Windows at all.
MS can be so much better if they actually applied their research to Windows, its flagship product. But since Microsoft has already accomplished its goal, have 90% worldwide marketshare on operating systems, I guess it can get away with incremental improvements every half decade or so. It's not that Microsoft doesn't innovate, look at the research. That's innovation. It's that Microsoft doesn't want to capitalize its innovations and is content on sticking with its Windows (and Office) monopoly.
Project management is not a computer science skill; it's a software engineering skill (software engineering != computer science, although I feel, by experience, that software engineering skills are very important for computer scientists to learn). However, there are schools now that offer software engineering degrees which do explain some of the methodology and practices behind software engineering. Project management does fit in with a software engineering degree, exactly in the same way as project management techniques fit in with the traditional engineering disciplines.
I believe that for most industry positions in the software industry, a software engineering degree would be a better fit than a computer science degree, since the software engineering degree better prepares students for some of the conditions and practices that they'll see in industry. At my school, the software engineering degree also contains courses on teamwork and other engineering practices (sometimes alongside traditional engineers). A computer science degree, by definition, is geared toward people who want to become computer scientists. Computer scientists are focused more on research than software engineers, who are focused on building software. A computer scientist is to a software engineer as a physicist is to an electrical or mechanical engineer. Scientists are focused on expanding knowledge and doing research, whereas engineers are focused on applying scientific knowledge to practical uses.
Read the article again. The article talks about how IIT graduates are doing well in the industry because of their high quality of teaching. The main focus of the article, however, is on other Indian universities, not IIT (which is one of the best schools in the world).
There is only one problem with the Libertarian Party: much of the official party's platform needs to be toned down from some of the borderline anarchocapitalistic stances that the Party takes on certain issues. I feel that the Libertarian Party can make a big splash as a party that stresses the synergy of protecting civil liberties and fiscal conservatism, rather than a party that wants to turn the USA into Ancapistan overnight. Plenty of fiscal and economic conservatives disgruntled about the direction that the Republicans are taking, and aren't fond of the growth of social conservatism would be very interested in a moderate Libertarian party. Liberals who don't like high taxes and heavy government spending, and are disillusioned with the idea that the government would fix economic issues, would also like the party. (I am personally an ex-liberal who turned into somewhat of a libertarian).
The Art of Computer Programming is about algorithm analysis and development, not about computer architecture (caching) or software development (profiling). TAOCP, combined with Concrete Mathematics, the MIT algorithms book, and a good text on combinatorial algorithms should be more than enough to satisfy the needs of a computer scientist or seasoned software developer who needs all of the good algorithms that he or she can get. TAOCP isn't a book about all of computer science, but it isn't trying to be one either.
K&R is a useful book for C reference; in fact, I consider it as an explanation to the C standard. I refer to K&R, not necessarily the standard, for reference whenever I am stuck on a certain C concept. (Compiler construction is different; you should follow the standard to the letter). There is a reason why the K&R book hasn't been updated since 1989; other than C99 extensions, the book is a very good explanation of C as is. I learned C largely from this book (although, to be fair, I used an O'Reilly book to learn the basics of C before taking a class that used C. The professor in that class didn't assign a book, so I sought out the best book according to ratings: K&R. My book is worn out out of 2 years of use, but I love it).
In California, taking a grocery cart out of the parking lot of a store is illegal. It is called shopping cart theft; you're only supposed to use the shopping cart in the store and within the parking lot. So, using a shopping cart outside of the store is not a legal option. Any other suggestions?
Well, I would argue that automobiles are a symbol of freedom. Automobiles allow us to get from door to door, without having to walk for over 5 minutes to a transit stop and without having to transfer buses a few times if you're going longer than a few miles. Automobiles allow us to travel anywhere we choose, instead of wherever the bus or train line goes. With an automobile, getting a load of groceries is easy. Try getting groceries without a car; you wish that you had one by the time you finally carry your groceries over (or give up and pay the taxi). Automobiles allow us to mind our own business. I don't want to listen to a passenger curse somebody out on a cell phone, or deal with some of the seedier characters on the bus or train (this is actually a common occurrence riding public transportation in Sacramento). Finally, with an automobile, we don't have to revolve our lives around other peoples' schedules. You can leave whenever you feel like it with a car. You can't do that with public transportation.
Am I saying automobiles are perfect? No. They are energy inefficient, for one, and I wish hybrid technology will further improve. Our roads are getting much more congested everyday, and our political leaders (in California) refuse to do anything about it (they have refused to build any roads on a large scale statewide since the Jerry Brown administration started in the mid-1970s. Don't get me started. To make a long story short, the Jerry Brown administration dismantled California's grand statewide freeway plans of the 1950s and 60s [which would have been completed by the 1990s], helped cancel many freeways in metro areas [Sacramento, Los Angeles, Bay Area except San Jose, and other areas], and left current construction projects abandoned. Thanks to his administration, and the lack of will from subsequent administrations to roll back his changes, California's highway system went from the best to the worst. Read more about it here.). However, we, as Americans, appreciate to mobility and convenience that we get from automobiles, and I do not feel that public transportation is as flexible or as convenient as automobiles are. We are willing to suck up congestion and higher gas prices in order to ride our automobiles.
As a road geek, prove this. Now, I know that roads cannot compete with bullet trains when it comes to speed and delivering passengers. (I'd like to see some more bullet train development in this country; California proposal of a train running from San Francisco to Los Angeles via the Highway 99 corridor is a start). However, I believe that roads, especially in the form of expressways and freeways, are still an important and optimal mean of transportation for distances ranging from 5 miles to 500 miles. Freight trains also cannot compete with trucks for getting goods from manufacturers, farms, and shipping ports to your local store. With freight trains, the goods still must make it from the train station to the store.
Besides, there is nothing like the free feeling that you get from driving a car or even riding along in one. Roads can take you almost anywhere. I love the freedom of travel that you get from riding on the many roads that are available. Riding in a car on a long road trip feels like an adventure; something exciting is about to occur. Riding in an Amtrak train (I've never rode a bullet train, so I can't talk about that experience), in my opinion, just don't give me that same exhilaration that riding in a car gets. It can sometimes be a close to grinding experience, especially when your trains get delayed (I'm an occasional Amtrak rider, so I had these experiences first hand). I also find certain roads to be beautiful achievements of civil engineering (go down to Southern California and see the freeways in Orange County; it's impressive). You get some emotion out of fellow road geeks. Now try to get that same emotion out of train structures.
To put it this way, trains are an important method of transportation, and bullet trains are unmatched in their speed and efficiency. But do not neglect the importance and utility of well designed roads. Roads are unmatched in the ability to be able to get from door-to-door without having to transfer, switch modes of transportation, revolve your life around a posted schedule, and other inconveniences.
Databases are a very important part of computer science. Databases are fundamental tools for many computer science (and other) research projects, and databases themselves are still being researched.
I still believe that a computer science and software engineering split is a great idea. In fact, at my school, the computer science department has two separate programs: computer science and software engineering. (A computer engineering degree is also offered, but it is its own department). Computer science is for people who want a more general exploration of computer science topics, software engineering is for people who want to know about the software engineering process, and computer engineering is a combination of computer science and electrical engineering.
Linux may win a war versus Microsoft on the server, but that is not where Microsoft is most powerful. Microsoft is still the 800 pound king gorilla of the desktop, and Linux still has a way to go before it unseats Microsoft. Heck, OS X has been out for over five years. Many users who have used both OS X and Windows claim that OS X is superior, and many have switched. However, OS X has barely pinched Microsoft, and Microsoft still enjoys 93% or so marketshare on the desktop (5% goes to Mac, and the rest of that is Linux, BSD, and other OSes).
The reason why Windows still hasn't been unseated is that too many people have software that is Windows-only. Businesses still rely on in-house Windows programs that were created by some programmer many years ago who is long gone, and cannot afford porting it to another operating system. Sometimes you'll see a business run an old Windows 3.1 or even DOS application since there is no replacement, and since it is good enough for them to not worry about porting it or creating a clone of it. Engineers aren't dropping AutoCAD anytime soon, and AutoCAD is Windows only. Engineers, being well known for their pragmatism, stick to Windows. Graphics artists on both the PC and the Macintosh who rely on Photoshop, Quark Express, and Dreamweaver are not going to move to Ubuntu and use the GIMP, Scribus, and nvu (yes, those open source products are good, but their commercial competitors are very good and are worth the $$$ that you pay for them; they end up saving you $$$ with their features and ease of use). And developers who want some food on their table better know something about Win32, .NET, and other Windows technologies. In most non-CS fields, you cannot avoid Windows in the professional world, and Windows has became a fact of life in many careers.
So, what is the open source community going to do about this? A great operating system with all of the bells and whistles isn't enough for most people. Once again, OS X is considered the best operating system by many people, but some applications haven't been ported yet (or won't be ported), which doesn't leave OS X as an option for those people. The open source community needs to start polishing up their offerings and get started on some new stuff (an AutoCAD replacement will get engineers off of Windows, for example). GIMP can use some improvement. OpenOffice should be more modular and faster. Dia needs to start looking like OmniGraffle or Visio. There needs to be some sort of OSS equivalent to Visual Basic (what I mean by that is ease of developing GUI applications). I recommend the same with other Linux applications.
Remember, the key to operating system adoption is applications. Look at MS-DOS, for example (back in its heyday). It was hard to use (compared to the Apple Macintosh at the time), very rudimentary (compared to other OSes in the 80s like Unix, NeXTSTEP, and VMS), and can only run one application at a time. But it ran Lotus 1-2-3 and WordPerfect, and it ran all of the other applications that business users wanted. Home users wanted the computers that business users had, so they got that too. Ditto for Windows 3.1. NeXTSTEP knocks the socks off of Windows 3.1. But who had the applications?
Your OS can be the easiest to use OS in the world. It can have microkernels with the best scheduling and load-balancing algorithms that exist. It can utilize all of the systems research published in the ACM and IEEE journals within the past two years. It can be so secure that it would be the envy of Homeland Security and would make Symantec and McAffe angry (they can't sell protection for it). It can even have a mass advertising campaign with beautiful angelic models praising the product. But if it cannot run the applications that they want, then it is just a waste of hard drive space and time as far as they are concerned.
I don't like that idea, because prospective computer scientists (not software engineers or computer engineers) would not have an undergraduate major reflecting their future career goals. The more theoretical of them (those going into algorithm development rather than systems) would be forced to become math majors, as that is the closest thing to theoretical computer science (and while math is a fun and great subject to learn, mathematics and computer science are different, although related). However, I believe that at some schools the computer science degree has tried a "one size fits all" approach, with mixed results. I believe that there should be four separate computing-related majors:
I prefer this system to getting rid of the general computer science degree. Keeping a general computer science degree for the future computer scientists of the world, while creating degree programs for those who want to specialize in software engineering, computer engineering, or business, better supports the diversity of career goals and academic goals of students interested in some computer-related field.
But that is not a computer science degree. Now, I believe that a computer science degree should show practical applications to real-world situations. However, theory should not be neglected. Courses such as algorithms, theory of computation, and computer architecture (which I noticed was omitted from your list; it can qualify as hardware, but it is important to know computer architecture, especially when doing systems programming) are very important for both software developers and computer scientists. GUI programming and client/server programming are not necessary in a computer science curriculum; GUI programming can possibly tie in to a HCI course, however.
Being taught all theory can be bad because you wouldn't know how the theories tie in to practical applications. But being taught all practice is just as bad, because you wouldn't know how all of the applications truly work. A good computer science program should have a balance of theory and practice.
Exactly, and I'm speaking as a libertarian. The problem with the Libertarian Party is that the official platform sounds like something that comes out of an anarchocapitalist book. Radical libertarianism and anarchocapitalism seems to cause more harm than good. I am not a supporter of the welfare state, but I do support some level of state-level assistance for the poorest of people. I happen to like anti-trust regulations (they keep the market free from the control of monopolies). I also like the Interstate Highway System (and state highway systems), and I would like to see more work done in my state (California) with freeway construction, the proposed bullet train, and other infrastructure improvements. I happen to agree with anti-discrimination laws and other civil rights laws. I believe in state-funded public education, with school choice and eventually vouchers. Those are some positions on issues where the Libertarian Party platform differs from my personal beliefs.
Personally, I wish the federal government would return to its Constitutional roots of limited government, fiscal responsibility, and civil liberties. The federal government has amassed far too much power in the past century. The Income Tax Amendment of 1913 and the Federal Reserve gave the federal government the income collecting mechanism needed to expand the government, and the government exploded in size during the FDR administration. The federal government hasn't been the same since. The original purpose of the United States of America was to focus on the states, not America. The power should return to local communities and states, not the federal government. This country wasn't called the United States of America for no reason.
On a state level, I would like to see more pragmatism instead of ideology. Speaking in terms of California politics, California needs to return to the days of the Pat Brown administration of the 1950s and 60s. Back then, California built the most impressive highway system in the world at the time, organized the higher education system (which features the top-notch UC system, the ubiquitous CSU system that has educated hundreds of thousands of Californians, and the community college system for everybody who wants a higher education), built the aqueducts, and did that all while still remaining quite fiscally prudent. California also had the best public school system in the country at the time. Nearly 50 years later, we Californians live in a shadow of those days. Our highways are still largely of 1960s vintage and are crumbing and in need of improvement, our great highway plans of the 50s are incomplete (freeway plans have been largely abandoned since the beginning of the Jerry Brown administration; don't get me started on him), our public schools are among the worst in the nation, and businesses are leaving California thanks to their sky-high taxes and the high cost of living due to lack of affordable housing (environmentalists stop developers from building on open space near the metro areas, so developers respond to them by building houses 30 miles up or down the road, resulting in crazy traffic congestion).
On a state level, I would like to see their government kind of like the Japanese government. The Japanese government is a model of pragmatism, in my opinion. They are able to fund quite a bit of services (such as highways, bullet trains, and even universal health care) while keeping taxes in check and by not running up huge budget deficits. The government also firmly believes in right-wing economics, such as privatization of certain services (universites have been partially privatized, the postal service has just been privatized, etc.), and they also believe in fiscal conservatism. The Japanese public K-12 school system is the envy of the world, and their universities are very good. On a state level, we need to be more like them.
As for the Libertarians, they need to be much more pragmatic, while at the same time not sacrificing the core of their principles: civil libertarianism fused with fiscal conservatism. A pragmatic version of libertarianism would almost certainly make inroads with much of the population segment that are not extreme leftists nor extreme conservatives.
I'll believe this when I see it.
Anybody can create a virus for OS X, and it can run perfectly. The biggest problem would be how it can be able to spread to other machines.
On Windows, it isn't viruses that plague Windows, but it is worms, spyware, and adware that affects that platform. All it takes to be infected with a computer virus on any platform is to not be vigilant about the data that you download. Being infected by spyware and adware, however, relies on the security of the browser, and being infected with a worm relies on the security of the operating system's Internet connectivity.
OS X remains relatively secure because its browser does not have hooks to the shell (unlike older versions of Internet Explorer, although I've read that Internet Explorer 7 has been decoupled from the shell), and because its Unix core isn't susceptible to worms (Unix has come a long way since the worm of 1988). OS X also has a firewall, although I just learned that it isn't enabled by default (but turning it on is easy; they should change the default in OS X 10.5).
A demo virus for OS X or Linux isn't news. No operating system can block the execution of a virus unless the operating system has a list of trusted applications that it knows are virus-free. An operating system can prevent worms with better security, and spyware can be prevented by using a secure browser, but viruses cannot be blocked from execution.
That is exactly my point. The Linux kernel hasn't changed too greatly in the past 5 years, but the desktop environments for Linux (and BSD) have improved greatly. Linux is harder to judge since there are so many separate components, so I focused on the desktops, since that is the most fair comparison to OS X and Windows.
Five years between operating system releases is a very long time in the computer industry. Look at how OS X improved in five years (from OS X 10.0 to OS X 10.4, with OS X 10.5 coming out sometime in the spring). Look at how KDE and GNOME have improved over 5 years. Look at some other 5-year periods of time in the computing industry. From 1991 to 1996, we went from DOS and Windows 3.0 to Windows 95 and Windows NT, and that is just on the MS side of things.
Five years without any changes other than security fixes and point updates to software is very sad. (Some would say six years if they believe that XP isn't that much of an improvement over 2000, but I'll use XP as a base here).
Am I the only person appalled by these web interfaces, or even web desktops, being referred to as operating systems? It is technically wrong by a large margin. An operating system is the interface between hardware and software that manages the resources of hardware. Web "operating systems" do not manage any hardware.
I find this usage appalling, and I hope that this terminology doesn't spread and dumb down the use of technical terms.
On most Cocoa applications (those that use APIs derived from NEXTSTEP; i.e., most applications that are from the NEXTSTEP days or are applications made 2001 and later) that utilize large toolbars, the toolbars can be fully customized and can be made smaller. The buttons won't be very small like the Office 97 buttons, but they won't be the larger default size, either. Carbon applications (those that use APIs derived from Mac OS 9 and below; i.e., most Mac applications that existed before OS X, such as MS Office and Photoshop) are based off of the older Mac OS 9 and below standard, which used small buttons for toolbars.
There are some Cocoa applications (Safari, Mail.app, TextEdit, Address Book, and a few others) that use small toolbar buttons rather than large ones. You should be fine. As for removing the icons altogether and replacing them with solely text, it depends on the application. Cocoa applications should be easier in this regard to be customized than Carbon ones (but, then again, I don't have any Carbon applications to test this out with, other than iTunes). I hope that your switch to the Mac goes well.
On my MacBook (1.83GHz Core Duo with 512MB RAM), I usually have the following applications open:
Depending on the task on hand, I may have a word processor, text editor, some X terminals and X11 applications (via a SSH tunnel to a school account where I do my CSC homework), iCal, or TeXShop open.
I also notice that the amount of windows that I have open is proportional to the speed and performance of the computer that I'm using. For example, when I was using my Duron 950MHz with 384MB RAM and either Windows XP or FreeBSD, I normally only had my web browser (Firefox or Konqueror), GAIM, my email client (Thunderbird or Kontact), and (on Windows) iTunes opened, and sometimes some terminals opened whenever I was doing CSC homework. I would run almost as many windows on that machine as my Mac, but I would be a bit more cautious about performance. On my 266MHz Pentium II laptop with 64MB RAM running FreeBSD, I normally stick to just one application (usually my web browser, Opera, and even that one application can be demanding. (Running some X terminals, gvim, Opera, GAIM, and another GTK app on that machine is an exercise of patience whenever switching between applications, and I couldn't even begin describing the horrors of using OpenOffice on that machine).
Innovations such as Expose and (on KDE and WindowMaker, the desktops that I used on my BSD machines) virtual desktops make using multiple windows much easier. I tried not to have more than 4 or 5 windows opened on my non-Mac machines because there is no easy way of finding all of them (although virtual desktops do help with organizing categories of windows). Because of Expose, I am able to have up to 15 windows on my screen (I've worked with that many windows once) on my Mac and still work very efficiently and be productive without being lost. Bringing virtual desktops to OS X (something that I miss from my BSD experience) and giving virtual desktops the power of Expose will make my Mac even better to use.
Yes. I downloaded the Office 2007 beta back in May and played around with the ribbons interface. The ribbons interface is very good if you are a beginner and you are trying to find options. However, I am far from an Office beginner; I've used Office since the Office 97 days. I know how to customize the toolbar so that way every option that I've needed was accessible from one click from the toolbar instead of multiple clicks from menu options. The ribbons, however, made many things that I was once able to do with one click now take two or even three clicks to do. It takes up a lot of space on the screen, looks ugly on XP and doesn't conform to any pre-Vista Windows standards (but it blends in with Vista very well), and is uncustomizable and can't be removed (the "if you don't like it, tough" approach). It seems to get in the way of my work, and seems to be a much clunkier solution than the old toolbars/menus or even the Inspector approach on OS X.
My biggest problem with ribbons is that it dumb downs Office. Office is a professional tool used to make spreadsheets, documents, presentations, databases, and more. It is inherently complex, kind of like Photoshop, Quark Express, and other complex tools. In the old versions of Office, getting used to the toolbar options and menus may be a bit tough at first, but after finding the options (which is just a few days), you will be productive easily. You can even customize Office to fit your needs and for you to access your most commonly-used features in just one click. Office 2007 forces people to stay beginners forever. Everything requires extra clicks to do compared to Office 97, 2000, XP, and 2003. There is no way to make them one click options, short of memorizing the character shortcuts (which the ribbons do not show, unlike the menus, which show keyboard shortcuts). Office is inherently complex, like Photoshop. Yet if Photoshop's or Quark Express's interface dramatically changed to be beginner-friendly yet professional user-hostile, you'd see a riot in every design company's office.
Perhaps the ribbon interface is a better experience on Vista, where it blends in with the rest of the interface a bit more. Still, I do not like Microsoft's approach to usability. MS's approach tends to dumb down the interface to the point that it is usable for novices and other people who don't have much computer experience, but very difficult to use for people who know how to use a computer. Apple doesn't dumb down anything; it balances creating a usable OS while making it flexible and usable for advanced users at the same time. You can make your interfaces easy to use without alienating experienced users. That's my problem with Office 2007. It makes things easy for beginners at the expense of experienced users.