Playing devil's advocate, I would have to say that HP dishonors warranties because sometimes it is too difficult to determine whether a computer issue is hardware or software related. Which makes business sense if you think out it, HP can't possibly train their technical support staff on how to navigate every possible operating system and every possible problem that can occur within those systems. Although a keyboard not functioning may seem like a stupid hardware problem to you and I, a mis-installed or corrupted OS can mimic lots of hardware problems. I know, I know... It's sometimes difficult for Linux people to comprehend concepts such as financially feasible or what makes business sense, but at the end of the day, companies like HP don't base decisions on geeky customers looking to run their favorite operating system but on something that begins with the character '$'. Sorry.
BTW, Is it me or is Slashdot's CAPTCHA images getting a little too carried away?
What the state of Connecticut should do is hand out electronic scales that connects to the computer and if the kid is too fat he can't log in. Or the kid can log in but he can only visit www.weightwatchers.com.
I call this technology "FatAss.NET". Not a bad idea, ehhh???
Two words, Anonymous Coward. an Anonymous Coward who obviously has reading problems. Or better yet, comprehending what others post. If anything, your an example of exactly the kind of schmucks that are out there ruining Linux, not helping it.
I think the problem is not with Linux itself (the kernel) but with the concept of open source. Not to say open source software is poor quality by any means, but that the most of developers involved in creating this type of software seem to put very little emphasis on usability for non-geeks. To see this in Ubuntu, all you have to is go to the synaptic package manager and bring up any package. Most require two, five, ten, sometimes twenty different dependencies in order to function. Most people don't understand (or don't care to understand) why AbiWord requires abiword-common, libart-2.0-2 and libatk1.0-0 and that's if you have enough bandwidth to download them in a reasonable time period. Although the package manager does a descent job of locating and installing all of the required packages; uninstalling them without errors is hit and miss at best and then your stuck hand editing configuration files so that dpkg can can continue removing the remaining packages. NO ONE WANTS TO DO THIS!! On any modern day operating system, libraries (or any dependencies) should not be of concern to everyday users and so anything that begins with "lib" should not have to be dealt with by the user. Plain and simple. This is excatly why Linux will never be taken seriously by desktop users. Fancy icon and screen saver graphics and are not enough. Now it's time for some serious usability overhaul.
It's all been done before in Linux. They tried moving Delphi to Linux (Kylix) and what happened?? Nada! And Delphi is a great language and would have had a big enough impact if the underlying UI system wasn't so caked. VB in Mono is not going to make a difference. We've also seen TCL/TK, Perl/TK, QT, GTK, Glade, etc. And now comes Mono. Yet, still hardly anyone wants to develop Linux GUI apps, and I can't say I blame them. Command line apps are still a dime a dozen in Linux. But GUI apps?? Too much work! Why? The way I see it, regardless of the library or language, there are way too many levels of libraries the user needs to worry about installing and configuring before he/she can even attempt use the application. No one wants to deal with this, which is why Linux still sadly remains a mostly server only system. Plain and simple. Truth is, the XWindows layered design has failed Linux in taking on the proprietary systems and will continue to do so regardless of whatever language, library or development environment you develop in. I've been using and developing in Linux and Windows for a long time and I can say with utter certainty that years will pass and we'll still be having these discussions as long as the GNU/Linux philosophy is to cake on layers of complex GUI libraries. Sadly, adding a bastardized version of VB as a GUI development tool in Linux doesn't change the facts and is destined to suffer the same fate as Kylix. Just no way to put a nice face on it folks. Don't get me wrong I like Linux a lot, but I'll stick with developing GUI applications in the Windows environment for now.
I think it's great that you want to start a new career. There are two basic ways to go about it. Working for a company or working on your own. Working on your own is not always that easy because it's sometimes hard to find work. You can always advertise in the local paper, perhaps building websites or creating databases for local small businesses. There are also websites I've seen that let you bid on development jobs, although I have no experience with this. Bottom line you'll have to keep at it and have patience.
As for companies hiring developers, it's unfortunate that you'll hear a lot of people complain that there is no work because of outsourcing and that the Indians are doing everything for 10 bucks and hour. I'm not saying outsourcing isn't a problem but lets face it, today's companies have smartened up quite a bit. What I mean is, they're not willing to hire programmers who've just learned to code last last year by buying a Microsoft Visual Studio book and following examples. That was the 1990's when every schmuck started a company and paid programmers six figures hoping to make millions on Java applets. The days of VB programmers getting 100k/year are long over. Today you gotta be good, you gotta be flexible and you gotta be your own best salesman. Also understanding the company you work for and how your efforts will fit into their business, in my view, is very important to getting ahead. Oh, I almost forgot. Stay away from recruiters if you can help it. Recruiters are mostly a bunch of phony dick heads that exist to make money by wasting your time. Trust me, try to meet the potential employer on your own if possible. Good luck!
"Red Hat asserts that all the important security and hardware certifications would be invalidated on Oracle's offering."
Whoopdy do! Blah, blah, blah... Who do they think they are? Microsoft? Just remember, this is what you get folks for paying $$ to certify for a corporate controlled open source product. Never know when someone else is gonna move right in. Helk, everyone shares the same source code! No one's land locked. I guess trying to become the Microsoft of Linux proves to be a bad idea after all.
I figured the "do-it-yer-selfers" would be such an insignificant base of customer that Microsoft would bypass them entirely and force them to re-register (or re-purchase) the product if the hardware changed. Perhaps they are using this "out of the goodness of their hearts" excuse as a way to hide the real reason. They don't want to receive thousands of wasteful support calls every time a guy named Chip changes his network card.
Well Cliff, you've discovered something thats been irritating me (as well as many other multi-platform developers) for years. We all agree Linux (the kernel) is a great system. However, a lot of GNU tools that accompany the OS are unfortunately lacking in ease of use, and as such has actually hurt the developer's ability to create user friendly software for Linux. User friendly software requires not just good programming skills but also requires that the developer can spend time focusing on writing his software and not on configuring his development tools. Lets face it, lots of developers hate Microsoft (mostly for political reasons), but fail to realize just how sophisticated and helpful their development tools really are. Moreover, its because of these great tools that lots of high quality GUI software has been written in Windows and not in Linux.
Lets look at another industry for example. Take a car mechanic. Go to any shop today and you'll rarely see a mechanic slide under a car anymore. Today hydrolic lifts are used. Or will you not see a hand socket being used to remove your cars tire lugs. Instead, the mechanic will have use a pneumatic (air) gun to remove the lugs. Now what you don't see in this industry, are mechanics still using the hand wrench and the floor dolly telling other mechanics that they are not knowledgable because they don't use the less sophisticated tools. In fact, pretty much all mechanics want to use these tools. Perhaps the cost of a lift was the prohibiting factor not a "tool religious war". Why I mention this is that unfortunately, in the unix development industry, most developers don't see the merit in using anything other than archaic console development tools to write GUI applications perhaps because it makes them less geeky? Who knows.
At first some have broken down and developed third rate GUI applications in X windows (such as xxgdb, DDD, graphical emacs and vi tools), but no where near the likes of a Visual Studio. When people got tired of these horrific screen scraping abortions next came, what I like to call the "somewhat functional" IDEs (KDevelop and glade), still no where near the likes of a Visual Studio. I say somewhat functional because they still require about 6 to 10 different libraries to be downloaded, installed and hand configured to even be able to compile a simple "hello world" application! After this, open source GUI developers began to question the languages of which they were developing. C and C++. The mindset now became, "What if we were to develop GUI apps with a simpler language using the same GUI libraries we used for C/C++ development. And so became the bindings war! Perl-TK, GTK-Perl, pyGTK, pyQT, PHP-GTK, PHP-QT, QT#, GTK#, tcl-GTK... You name the language and it has a GUI development system for it and a group of developers who SWEAR by it. Of course you still have to pray your distro comes with these languages and all it's libraries installed. Otherwise, your back to the same problem you had the C/C++. Besides, none of the above mentioned languages still provides a useful IDE.
"Ok, enough already!" you say. "So what constitutes a good IDE?" Here are a just a few of the features an IDE (in the year 2006) should have regardless of OS.
1. Editor supporting syntax highlighting and code folding 2. GUI builder with a toolbox and generation of event code when a control is double clicked 2. Automatic makefile generation. 3. Full code completion support for all GUI and unix system libraries (including completion for classes and functions you create) 4. Code completion that includes a parameter list that indicates the parameter you're cursor sits at 5. Built in visual debugger with a tree-like hierarchy displaying members inside classes and structs 6. Ability to drag and drop variables to watch windows 7. An immediate window to run code and execute expressions while debugging 8. An editor that lets you find variable and class declarations by clicking on a variable 10. Mouse-over variabl
> On most OSes, shared memory is a precious resource and requires either static, > compile-time tunables, or run-time tunables that are difficult to juggle. Shared memory is > hard to port between OSes because nobody supports it quite the same way.
Tell me this is not what you wrote. If it's not then it's time to change your/. password. Shared memory is not all that difficult to port between OS's if you coded correctly to start with. I've done it many times before, so before having a hissy fit know what you are talking about. Finally, I don't think you needed to "qualify" your above sentence anymore than you already did.
> I'm not even pretending to extend this to a generalisation; you inferred that because I > didn't qualify *every* single one of my f-ing sentences. That's pretty pedantic bud.
Ok, Ok. So next time be more careful when you type, and be nice to people who call you on it. Besides your use of the word "f-ing" needs two more dashes. (f---ing)
> Postgresql would be a real bitch to install and support, and therefore it's not worth the > time and effort it would require. People in my situation (with large-scale requirements) > would face the same problem.
Fair enough, but I happen to know people who use postgresql in large scale situations and don't share your problem. Oh well.
> Pay attention, dammit.
Thats the spirit! Now, channel that anger into solving your database troubles.:)
Years ago some people made a living sharpening knives. That is until someone came out with the knife that never needed sharpening, and those guys are now in other businesses or not in business anymore. Thats capitalism. People tend to forget that Microsoft, indirectly, helped create lots of software industries (and made some people very rich) as a by product of it's own software's limitations. Perhaps we all remember QEMM or Quarterdeck Expanded Memory Manager for DOS to help ease the famous 640K memory limit. How many people do you know that still use this product?
Believe it or not, it takes a tremendous amount of engineering and time to create a software system that can withstand the kinds of abuse that Microsoft Windows has been putting up with in the last 6 years. All kinds of new ways of hacking (ie. spyware, phishing schemes, spam, etc) have come along and have been pretty much directed at one system. Hopefully, this new version of Windows will be able to counteract these problems, saving money and time for businesses that rely on Windows. If all this means some companies will be forced to come up with new ideas or go out of business well then thats unfortunate, but thats also business.
> On most OSes, shared memory is a precious resource and requires either static, compile-time > tunables, or run-time tunables that are difficult to juggle. Shared memory is hard to port > between OSes because nobody supports it quite the same way.
You fail to mention that shared memory is by far the fastest form of data sharing between processes, no matter what the OS. With shared memory, (or memory mapped files in the case of Win32), there is no need to make expensive system calls to the OS once the memory is in place. The only expense is in syncronizing between reads and writes using semaphores. But even then, these operations are not really expensive for the CPU. Moreover, shared memory code is not all that difficult to port to other systems because the concept of memory is the same regardless. Pointers are pointers. Although the functions to create the memory are different between Unix and Win32 environments. Still most of the fancy code is done with standard pointers, and if coded with care (ie, using sizeof) then the software should be pretty portable.
> So now we have to recompile our kernels just to support Postgres? That is not cool,
No, only if you want to support a huge database (ie. huge table sizes). However, all systems have limits. Maximum number of open files, maximum concurrent processes. This is exactly the reason why software like PostgreSQL was designed to scale to any system's maximum limits. If the OS can support a TB of shared memory, PostgreSQL should use it, and I bet this is one of the reasons why Sun is including it in their Solaris 10 system, because Sun systems support huge amounts of memory.
> I'm sorry, but it's just not practical to support Postgres without extensive tuning and configuration
Extensive tuning? Thats news to me. I'm sorry my friend but I have to totally disagree there.
PostgreSQL is a fantastic free database. I've been using it for 5 years solid. I've run it on Linux and FreeBSD and it never once crashed. Plently of other people who have used this database will tell you the same thing. Sun even decided to include Postgres standard in it's Solaris 10 product. So whats the problem? The problem for PostgrSQL is that fact that it's counterpart, MySQL is mentioned in countless books, magazines and articles, surpasing PostgreSQL by a mile.
Why is this? Perhaps it's because most people who work with databases these days do so within a web server type environment. Most books and tutorials on web development seem to feature MySQL exclusively and since lots of aspiring web developers read these books, it's the database they've come to know by default. There's even the L.A.M.P acronym. Although I personally prefer, F.A.P.P (FreeBSD, Apache, PostgreSQL, PHP). But that doesn't sound so good.
FreeBSD suffers from the same problem. When people think of Open Source operating systems, they naturally think of Linux, not BSD. BSD has been around a long time and is extremely stable. Some even go so far to say that FreeBSD is more stable than Linux. So why is Linux a household name? Perhaps, it's the story of the underdog from Finland competing with the all mighty evil wizard from Redmond that makes it so a compelling? Who knows.
As a business owner and a computer engineer, I'm glad that Microsoft FINALLY took the idea of lesser privileged accounts seriously. It is true that they do allow a limited account, but it's not that customizable and lots of software simply doesn't work well under it. So people are forced to stick with running as the "dangerous" admin user. And this is all if the user knew how to create a limited account in the first place! It's simply a security joke. Unix was built on this concept since it's inception over 30 years ago and has proven quite effective ever since. In fact, the other day I was asked to install a printer driver into one of my customer's Macs, I was pleasantly surprised to see a root password dialog appear before the driver could be installed. In 2006, thats the way it should be!
Personally, I never understood why Microsoft waited all this time. You could patch the system till the cows come home but if your always giving the user full control of a system, then you're also giving viruses and spyware those same privileges! It's just that simple! Perhaps, their view in earlier years was that forcing mom and dad to remember passwords would piss them off? In 1985, I could understand that view. But in the connected world we live in now, I don't buy it. Now, if Microsoft implements this feature correctly, and lesser privileged accounts are created by default (out of the box), then I don't see a reason why virus and spyware attacks shouldn't be cut in half.
AOL??? It amazes me that to this day, in 2006, people still unfortunately require the services of this useless ISP. As far as I'm concerned, AOL is the appendics of the Internet. The Internet training wheels of the 1990's if you will. Their software has, and always will be, nothing but a controlled adware monstrosity used religiously by fruity moms and dads who never knew that right clicking the mouse brought up a fucking menu. And then we all wonder why our president has to go on national television to tell us that countries like India and China are kicking our ass in technology. There are only two things AOL ever did successfully, and that was give computer techs plenty of practice reinstalling Windows and waste a shit load of plastic on advertising CDs that every month offered 2^32 power more free, useless, online minutes than the previous month. Far as I see it, AOL is a sick joke and an embarrassment!
Yes, it's good for kids to learn how to use computers. BUT, the problem is that computers make things too easy and what happens is that the basics like simple mathematics and penman ship start to suffer. I remember, walking into a Boston Chicken restaurant and the electronic cash register had failed, so the young girl behind the counter had to figure out my change in her brain. She felt embarrassed and asked the manager to help her. This is growing a problem in our society. Computers are great tools, but instead of augmenting people's abilities, they are giving people a false sence of knowledge. I call this the "Value Menu Syndrome", because all these kids have to do is press 1 for a Big Mac and 2 for a double cheese burger. The math has been eradicated.
This is part of the reason (I feel) that people in poorer countries are starting to beat out America's kids in math and science simply because the lack of constant access to computers allows kids to be more creative by forcing them to exercise their brains more.
As a computer engineer myself, I feel it would be more beneficial if kids studied computers a little later in their youth (ie. 16+) and learn the basics early. The Internet is a great and convenient way to find information, but not all information found on it is true. It's easy for any shmuck to post a web page and claim his/her material is true or reliable when it fact it may not be. It's also very easy for kids to find hate and other forms of racist and highly opinionated material that can alter their sence of the "real" world. It's best for kids to remain somewhat innocent and use their brains to solve problems, then when the time comes show them the computer which will help them solve problems faster and easier.
People who are computer savvy and malicious are always going to try to either attach some device or use a CD burner to steal information. This is where corporate culture comes into play by first, not allowing people to bring in and hook up any electronic devices to their computers and second, by locking down a machine enough where the OS will require a password to access ports such as USB. Unix can already do this and the next version of Windows, I believe, will have this capability as well.
have come a long way. Personally, I'm not a huge fan of open source, but for some things like databases and web servers, you can't deny that it has left it's mark in a big way. Now, I'm not very familiar with DB2 or SQL Server, so I won't comment on their features or lack of, but I've been introducted to PostgreSQL a few years ago, been using it ever since and I just can't see why (at least for small to medium sized applications) it wouldn't work just as well as an SQL Server or DB2. Postgres is awesome. You can even get commercial support for Postgres. In fact, I just heard that Sun is now including it in it's Solaris 10 system. Thats good enough for me. I don't see why one would shell out $$ for a database anymore, except for support.
It's not that expensive to buy. You can construct a GUI in minutes and access to the Windows API is simple through the API Viewer. VB 6 is still a very viable solution to writing quick apps because the language is simple, you don't have to learn a big API (other than WinAPI of course), you don't need a big runtime to distribute or download and there are tons and tons of web sites, articles and code snippets available.
> Is the slightly more expensive Mac really that more expensive over time? I think not.
Mac's are not slightly more expensive. I can buy a $8000 PC too, that will blow away a Mac, but that doesn't mean I need to spend $8000.
> that is now 6 years old but running MacOS X 10.4.4 and doing it well enough that I do not feel that I need to upgrade the machine yet.
The same can be said for PC. There are plently of PC owners who have their machines for the same amount of time and don't feel they need to upgrade.
> The fact that Apple uses higher quality components in and of itself justifies a price that is higher than that of a PC
Sure maybe a couple hundred bucks difference. I could see that, but not twice the price!
Look, I'll bet any money that Apple's OS will run just as fast and just as stable on any modern day PC and I'm not alone. All your dealing with here is a BSD OS with a really nice GUI. You certainly don't need to buy any "special" hardware from Apple for that. It's all marketing and appealing to people that love hearing all the fancy buzz words like hyper-threading and dual core.
And obvious. Who really cares about Mac hardware? It's all the same components anyway. Is Tropicana OJ any different than the supermarket brand? No, same orange growers different labels and price tags. So why pay $1200 for Apple's hardware when a $500 Dell will do just fine? And don't tell me that Apple uses flux capacitors and so thats why it's twice the price! It's all bullshit. If Linux (an open source product) can run on multitudes of hardware, why not the Mac? Even Sun has the decency to let you run Solaris x86 on ANY PC you want.
Because Apple loves to play head games. They get you to believe that their precious hardware is somehow different than the rest of the world's simply because the box is translucent, fashionable and doesn't need a second mouse button. My PC doesn't need to be fashionable. Hey Jobs! It's obvious by now that you REALLY want people to move to Mac OS. Obvious enough that you moved your flagship system to the same architecture as your arch rival. So why don't you just sell me the damn OS (you know you want to) and let me pick my own hardware buddy! This ain't AS/400. Stop with the games. I'll make my own Apple experience with my own hardware. Capiche?
Slashdot Mirror
Playing devil's advocate, I would have to say that HP dishonors warranties because sometimes it is too difficult to determine whether a computer issue is hardware or software related. Which makes business sense if you think out it, HP can't possibly train their technical support staff on how to navigate every possible operating system and every possible problem that can occur within those systems. Although a keyboard not functioning may seem like a stupid hardware problem to you and I, a mis-installed or corrupted OS can mimic lots of hardware problems. I know, I know... It's sometimes difficult for Linux people to comprehend concepts such as financially feasible or what makes business sense, but at the end of the day, companies like HP don't base decisions on geeky customers looking to run their favorite operating system but on something that begins with the character '$'. Sorry.
BTW, Is it me or is Slashdot's CAPTCHA images getting a little too carried away?
What the state of Connecticut should do is hand out electronic scales that connects to the computer and if the kid is too fat he can't log in. Or the kid can log in but he can only visit www.weightwatchers.com.
I call this technology "FatAss.NET". Not a bad idea, ehhh???
Two words, Anonymous Coward. an Anonymous Coward who obviously has reading problems. Or better yet, comprehending what others post. If anything, your an example of exactly the kind of schmucks that are out there ruining Linux, not helping it.
I think the problem is not with Linux itself (the kernel) but with the concept of open source. Not to say open source software is poor quality by any means, but that the most of developers involved in creating this type of software seem to put very little emphasis on usability for non-geeks. To see this in Ubuntu, all you have to is go to the synaptic package manager and bring up any package. Most require two, five, ten, sometimes twenty different dependencies in order to function. Most people don't understand (or don't care to understand) why AbiWord requires abiword-common, libart-2.0-2 and libatk1.0-0 and that's if you have enough bandwidth to download them in a reasonable time period. Although the package manager does a descent job of locating and installing all of the required packages; uninstalling them without errors is hit and miss at best and then your stuck hand editing configuration files so that dpkg can can continue removing the remaining packages. NO ONE WANTS TO DO THIS!! On any modern day operating system, libraries (or any dependencies) should not be of concern to everyday users and so anything that begins with "lib" should not have to be dealt with by the user. Plain and simple. This is excatly why Linux will never be taken seriously by desktop users. Fancy icon and screen saver graphics and are not enough. Now it's time for some serious usability overhaul.
It's all been done before in Linux. They tried moving Delphi to Linux (Kylix) and what happened?? Nada! And Delphi is a great language and would have had a big enough impact if the underlying UI system wasn't so caked. VB in Mono is not going to make a difference. We've also seen TCL/TK, Perl/TK, QT, GTK, Glade, etc. And now comes Mono. Yet, still hardly anyone wants to develop Linux GUI apps, and I can't say I blame them. Command line apps are still a dime a dozen in Linux. But GUI apps?? Too much work! Why? The way I see it, regardless of the library or language, there are way too many levels of libraries the user needs to worry about installing and configuring before he/she can even attempt use the application. No one wants to deal with this, which is why Linux still sadly remains a mostly server only system. Plain and simple. Truth is, the XWindows layered design has failed Linux in taking on the proprietary systems and will continue to do so regardless of whatever language, library or development environment you develop in. I've been using and developing in Linux and Windows for a long time and I can say with utter certainty that years will pass and we'll still be having these discussions as long as the GNU/Linux philosophy is to cake on layers of complex GUI libraries. Sadly, adding a bastardized version of VB as a GUI development tool in Linux doesn't change the facts and is destined to suffer the same fate as Kylix. Just no way to put a nice face on it folks. Don't get me wrong I like Linux a lot, but I'll stick with developing GUI applications in the Windows environment for now.
I think it's great that you want to start a new career. There are two basic ways to go about it. Working for a company or working on your own. Working on your own is not always that easy because it's sometimes hard to find work. You can always advertise in the local paper, perhaps building websites or creating databases for local small businesses. There are also websites I've seen that let you bid on development jobs, although I have no experience with this. Bottom line you'll have to keep at it and have patience.
As for companies hiring developers, it's unfortunate that you'll hear a lot of people complain that there is no work because of outsourcing and that the Indians are doing everything for 10 bucks and hour. I'm not saying outsourcing isn't a problem but lets face it, today's companies have smartened up quite a bit. What I mean is, they're not willing to hire programmers who've just learned to code last last year by buying a Microsoft Visual Studio book and following examples. That was the 1990's when every schmuck started a company and paid programmers six figures hoping to make millions on Java applets. The days of VB programmers getting 100k/year are long over. Today you gotta be good, you gotta be flexible and you gotta be your own best salesman. Also understanding the company you work for and how your efforts will fit into their business, in my view, is very important to getting ahead. Oh, I almost forgot. Stay away from recruiters if you can help it. Recruiters are mostly a bunch of phony dick heads that exist to make money by wasting your time. Trust me, try to meet the potential employer on your own if possible. Good luck!
Someone's been playing too much Splinter Cell at Samsung perhaps? Trust me, all you need to do is throw a flare to defeat the robot gun.
"Red Hat asserts that all the important security and hardware certifications would be invalidated on Oracle's offering."
Whoopdy do! Blah, blah, blah... Who do they think they are? Microsoft?
Just remember, this is what you get folks for paying $$ to certify for a corporate controlled open source product. Never know when someone else is gonna move right in. Helk, everyone shares the same source code! No one's land locked. I guess trying to become the Microsoft of Linux proves to be a bad idea after all.
To all my Linux brothas and sistas, we don't need no fancy graphics pretendin to make our community betta!
So what is it we need?????
"BETTER DEVELOPMENT TOOLS!!"
Louder!! The Gods of Linux can't hear ya!
"B E T T E R... D E V E L O P M E N T... T O O L S !!"
One more time people!!!
"B E T T E R... D E V E L O P M E N T... T O O L S !!"
HALLELLUYA!!!!!!!! THE PENGUIN HAVE MERCY!!!
I figured the "do-it-yer-selfers" would be such an insignificant base of customer that Microsoft would bypass them entirely and force them to re-register (or re-purchase) the product if the hardware changed. Perhaps they are using this "out of the goodness of their hearts" excuse as a way to hide the real reason. They don't want to receive thousands of wasteful support calls every time a guy named Chip changes his network card.
Well Cliff, you've discovered something thats been irritating me (as well as many other multi-platform developers) for years. We all agree Linux (the kernel) is a great system. However, a lot of GNU tools that accompany the OS are unfortunately lacking in ease of use, and as such has actually hurt the developer's ability to create user friendly software for Linux. User friendly software requires not just good programming skills but also requires that the developer can spend time focusing on writing his software and not on configuring his development tools. Lets face it, lots of developers hate Microsoft (mostly for political reasons), but fail to realize just how sophisticated and helpful their development tools really are. Moreover, its because of these great tools that lots of high quality GUI software has been written in Windows and not in Linux.
Lets look at another industry for example. Take a car mechanic. Go to any shop today and you'll rarely see a mechanic slide under a car anymore. Today hydrolic lifts are used. Or will you not see a hand socket being used to remove your cars tire lugs. Instead, the mechanic will have use a pneumatic (air) gun to remove the lugs. Now what you don't see in this industry, are mechanics still using the hand wrench and the floor dolly telling other mechanics that they are not knowledgable because they don't use the less sophisticated tools. In fact, pretty much all mechanics want to use these tools. Perhaps the cost of a lift was the prohibiting factor not a "tool religious war". Why I mention this is that unfortunately, in the unix development industry, most developers don't see the merit in using anything other than archaic console development tools to write GUI applications perhaps because it makes them less geeky? Who knows.
At first some have broken down and developed third rate GUI applications in X windows (such as xxgdb, DDD, graphical emacs and vi tools), but no where near the likes of a Visual Studio. When people got tired of these horrific screen scraping abortions next came, what I like to call the "somewhat functional" IDEs (KDevelop and glade), still no where near the likes of a Visual Studio. I say somewhat functional because they still require about 6 to 10 different libraries to be downloaded, installed and hand configured to even be able to compile a simple "hello world" application! After this, open source GUI developers began to question the languages of which they were developing. C and C++. The mindset now became, "What if we were to develop GUI apps with a simpler language using the same GUI libraries we used for C/C++ development. And so became the bindings war! Perl-TK, GTK-Perl, pyGTK, pyQT, PHP-GTK, PHP-QT, QT#, GTK#, tcl-GTK... You name the language and it has a GUI development system for it and a group of developers who SWEAR by it. Of course you still have to pray your distro comes with these languages and all it's libraries installed. Otherwise, your back to the same problem you had the C/C++. Besides, none of the above mentioned languages still provides a useful IDE.
"Ok, enough already!" you say. "So what constitutes a good IDE?"
Here are a just a few of the features an IDE (in the year 2006) should have regardless of OS.
1. Editor supporting syntax highlighting and code folding
2. GUI builder with a toolbox and generation of event code when a control is double clicked
2. Automatic makefile generation.
3. Full code completion support for all GUI and unix system libraries (including completion for classes and functions you create)
4. Code completion that includes a parameter list that indicates the parameter you're cursor sits at
5. Built in visual debugger with a tree-like hierarchy displaying members inside classes and structs
6. Ability to drag and drop variables to watch windows
7. An immediate window to run code and execute expressions while debugging
8. An editor that lets you find variable and class declarations by clicking on a variable
10. Mouse-over variabl
> On most OSes, shared memory is a precious resource and requires either static,
/. password. Shared memory is not all that difficult to port between OS's if you coded correctly to start with. I've done it many times before, so before having a hissy fit know what you are talking about. Finally, I don't think you needed to "qualify" your above sentence anymore than you already did.
:)
> compile-time tunables, or run-time tunables that are difficult to juggle. Shared memory is
> hard to port between OSes because nobody supports it quite the same way.
Tell me this is not what you wrote. If it's not then it's time to change your
> I'm not even pretending to extend this to a generalisation; you inferred that because I
> didn't qualify *every* single one of my f-ing sentences. That's pretty pedantic bud.
Ok, Ok. So next time be more careful when you type, and be nice to people who call you on it. Besides your use of the word "f-ing" needs two more dashes. (f---ing)
> Postgresql would be a real bitch to install and support, and therefore it's not worth the
> time and effort it would require. People in my situation (with large-scale requirements)
> would face the same problem.
Fair enough, but I happen to know people who use postgresql in large scale situations and don't share your problem. Oh well.
> Pay attention, dammit.
Thats the spirit! Now, channel that anger into solving your database troubles.
Years ago some people made a living sharpening knives. That is until someone came out with the knife that never needed sharpening, and those guys are now in other businesses or not in business anymore. Thats capitalism. People tend to forget that Microsoft, indirectly, helped create lots of software industries (and made some people very rich) as a by product of it's own software's limitations. Perhaps we all remember QEMM or Quarterdeck Expanded Memory Manager for DOS to help ease the famous 640K memory limit. How many people do you know that still use this product?
Believe it or not, it takes a tremendous amount of engineering and time to create a software system that can withstand the kinds of abuse that Microsoft Windows has been putting up with in the last 6 years. All kinds of new ways of hacking (ie. spyware, phishing schemes, spam, etc) have come along and have been pretty much directed at one system. Hopefully, this new version of Windows will be able to counteract these problems, saving money and time for businesses that rely on Windows. If all this means some companies will be forced to come up with new ideas or go out of business well then thats unfortunate, but thats also business.
> On most OSes, shared memory is a precious resource and requires either static, compile-time
> tunables, or run-time tunables that are difficult to juggle. Shared memory is hard to port
> between OSes because nobody supports it quite the same way.
You fail to mention that shared memory is by far the fastest form of data sharing between processes, no matter what the OS. With shared memory, (or memory mapped files in the case of Win32), there is no need to make expensive system calls to the OS once the memory is in place. The only expense is in syncronizing between reads and writes using semaphores. But even then, these operations are not really expensive for the CPU. Moreover, shared memory code is not all that difficult to port to other systems because the concept of memory is the same regardless. Pointers are pointers. Although the functions to create the memory are different between Unix and Win32 environments. Still most of the fancy code is done with standard pointers, and if coded with care (ie, using sizeof) then the software should be pretty portable.
> So now we have to recompile our kernels just to support Postgres? That is not cool,
No, only if you want to support a huge database (ie. huge table sizes). However, all systems have limits. Maximum number of open files, maximum concurrent processes. This is exactly the reason why software like PostgreSQL was designed to scale to any system's maximum limits. If the OS can support a TB of shared memory, PostgreSQL should use it, and I bet this is one of the reasons why Sun is including it in their Solaris 10 system, because Sun systems support huge amounts of memory.
> I'm sorry, but it's just not practical to support Postgres without extensive tuning and configuration
Extensive tuning? Thats news to me. I'm sorry my friend but I have to totally disagree there.
PostgreSQL is a fantastic free database. I've been using it for 5 years solid. I've run it on Linux and FreeBSD and it never once crashed. Plently of other people who have used this database will tell you the same thing. Sun even decided to include Postgres standard in it's Solaris 10 product. So whats the problem? The problem for PostgrSQL is that fact that it's counterpart, MySQL is mentioned in countless books, magazines and articles, surpasing PostgreSQL by a mile.
Why is this? Perhaps it's because most people who work with databases these days do so within a web server type environment. Most books and tutorials on web development seem to feature MySQL exclusively and since lots of aspiring web developers read these books, it's the database they've come to know by default. There's even the L.A.M.P acronym. Although I personally prefer, F.A.P.P (FreeBSD, Apache, PostgreSQL, PHP). But that doesn't sound so good.
FreeBSD suffers from the same problem. When people think of Open Source operating systems, they naturally think of Linux, not BSD. BSD has been around a long time and is extremely stable. Some even go so far to say that FreeBSD is more stable than Linux. So why is Linux a household name? Perhaps, it's the story of the underdog from Finland competing with the all mighty evil wizard from Redmond that makes it so a compelling? Who knows.
As a business owner and a computer engineer, I'm glad that Microsoft FINALLY took the idea of lesser privileged accounts seriously. It is true that they do allow a limited account, but it's not that customizable and lots of software simply doesn't work well under it. So people are forced to stick with running as the "dangerous" admin user. And this is all if the user knew how to create a limited account in the first place! It's simply a security joke. Unix was built on this concept since it's inception over 30 years ago and has proven quite effective ever since. In fact, the other day I was asked to install a printer driver into one of my customer's Macs, I was pleasantly surprised to see a root password dialog appear before the driver could be installed. In 2006, thats the way it should be!
Personally, I never understood why Microsoft waited all this time. You could patch the system till the cows come home but if your always giving the user full control of a system, then you're also giving viruses and spyware those same privileges! It's just that simple! Perhaps, their view in earlier years was that forcing mom and dad to remember passwords would piss them off? In 1985, I could understand that view. But in the connected world we live in now, I don't buy it. Now, if Microsoft implements this feature correctly, and lesser privileged accounts are created by default (out of the box), then I don't see a reason why virus and spyware attacks shouldn't be cut in half.
AOL??? It amazes me that to this day, in 2006, people still unfortunately require the services of this useless ISP. As far as I'm concerned, AOL is the appendics of the Internet. The Internet training wheels of the 1990's if you will. Their software has, and always will be, nothing but a controlled adware monstrosity used religiously by fruity moms and dads who never knew that right clicking the mouse brought up a fucking menu. And then we all wonder why our president has to go on national television to tell us that countries like India and China are kicking our ass in technology. There are only two things AOL ever did successfully, and that was give computer techs plenty of practice reinstalling Windows and waste a shit load of plastic on advertising CDs that every month offered 2^32 power more free, useless, online minutes than the previous month. Far as I see it, AOL is a sick joke and an embarrassment!
Yes, it's good for kids to learn how to use computers. BUT, the problem is that computers make things too easy and what happens is that the basics like simple mathematics and penman ship start to suffer. I remember, walking into a Boston Chicken restaurant and the electronic cash register had failed, so the young girl behind the counter had to figure out my change in her brain. She felt embarrassed and asked the manager to help her. This is growing a problem in our society. Computers are great tools, but instead of augmenting people's abilities, they are giving people a false sence of knowledge. I call this the "Value Menu Syndrome", because all these kids have to do is press 1 for a Big Mac and 2 for a double cheese burger. The math has been eradicated.
This is part of the reason (I feel) that people in poorer countries are starting to beat out America's kids in math and science simply because the lack of constant access to computers allows kids to be more creative by forcing them to exercise their brains more.
As a computer engineer myself, I feel it would be more beneficial if kids studied computers a little later in their youth (ie. 16+) and learn the basics early. The Internet is a great and convenient way to find information, but not all information found on it is true. It's easy for any shmuck to post a web page and claim his/her material is true or reliable when it fact it may not be. It's also very easy for kids to find hate and other forms of racist and highly opinionated material that can alter their sence of the "real" world. It's best for kids to remain somewhat innocent and use their brains to solve problems, then when the time comes show them the computer which will help them solve problems faster and easier.
People who are computer savvy and malicious are always going to try to either attach some device or use a CD burner to steal information. This is where corporate culture comes into play by first, not allowing people to bring in and hook up any electronic devices to their computers and second, by locking down a machine enough where the OS will require a password to access ports such as USB. Unix can already do this and the next version of Windows, I believe, will have this capability as well.
have come a long way. Personally, I'm not a huge fan of open source, but for some things like databases and web servers, you can't deny that it has left it's mark in a big way. Now, I'm not very familiar with DB2 or SQL Server, so I won't comment on their features or lack of, but I've been introducted to PostgreSQL a few years ago, been using it ever since and I just can't see why (at least for small to medium sized applications) it wouldn't work just as well as an SQL Server or DB2. Postgres is awesome. You can even get commercial support for Postgres. In fact, I just heard that Sun is now including it in it's Solaris 10 system. Thats good enough for me. I don't see why one would shell out $$ for a database anymore, except for support.
It's not that expensive to buy. You can construct a GUI in minutes and access to the Windows API is simple through the API Viewer. VB 6 is still a very viable solution to writing quick apps because the language is simple, you don't have to learn a big API (other than WinAPI of course), you don't need a big runtime to distribute or download and there are tons and tons of web sites, articles and code snippets available.
Yea, I agree. Just run my app first. Unless the app runs so slow it's not worth using.
I checked out this book at the local bookstore and it's definately worth buying. But you time to read it.
> Is the slightly more expensive Mac really that more expensive over time? I think not.
Mac's are not slightly more expensive. I can buy a $8000 PC too, that will blow away a Mac, but that doesn't mean I need to spend $8000.
> that is now 6 years old but running MacOS X 10.4.4 and doing it well enough that I do not feel that I need to upgrade the machine yet.
The same can be said for PC. There are plently of PC owners who have their machines for the same amount of time and don't feel they need to upgrade.
> The fact that Apple uses higher quality components in and of itself justifies a price that is higher than that of a PC
Sure maybe a couple hundred bucks difference. I could see that, but not twice the price!
Look, I'll bet any money that Apple's OS will run just as fast and just as stable on any modern day PC and I'm not alone. All your dealing with here is a BSD OS with a really nice GUI. You certainly don't need to buy any "special" hardware from Apple for that. It's all marketing and appealing to people that love hearing all the fancy buzz words like hyper-threading and dual core.
And obvious. Who really cares about Mac hardware? It's all the same components anyway. Is Tropicana OJ any different than the supermarket brand? No, same orange growers different labels and price tags. So why pay $1200 for Apple's hardware when a $500 Dell will do just fine? And don't tell me that Apple uses flux capacitors and so thats why it's twice the price! It's all bullshit. If Linux (an open source product) can run on multitudes of hardware, why not the Mac? Even Sun has the decency to let you run Solaris x86 on ANY PC you want.
Because Apple loves to play head games. They get you to believe that their precious hardware is somehow different than the rest of the world's simply because the box is translucent, fashionable and doesn't need a second mouse button. My PC doesn't need to be fashionable. Hey Jobs! It's obvious by now that you REALLY want people to move to Mac OS. Obvious enough that you moved your flagship system to the same architecture as your arch rival. So why don't you just sell me the damn OS (you know you want to) and let me pick my own hardware buddy! This ain't AS/400. Stop with the games. I'll make my own Apple experience with my own hardware. Capiche?