Call the SBC's DSL department and claim to be a friend "helping" someone install their DSL modem... but insist that you don't know the address or anything else. Be as dumb as possible on the phone. Get a little drunk if you can't be convincing.
Often, the customer service reps will read back the entire address, and sometimes, even the last for digits of the SSN. I found this out when I was ligitimately calling them because of a line problem.
I never had any problems adding service, removing service, or getting personal account information... all without identifying myself whatsoever. Need an address for a telephone number, call SBC and tell them you want DSL. The phone reps will "verify" your address by reading it back. Awesome, huh?
The ZCE hasn't earned even one extra dollar for me, but that's not why I have it. I needed wall art for my home office.
Certification is rather useless for ascertaining skill level, but does filter out the lowest classes of developer. I use certifications to skip really basic interview questions... for that, the ZCE is super useful. If you interview a lot, it's worth investing in.
My mother taught me how to read and my father taught me basic algebra. I started programming at age seven, already having the necessary skills. If I was forced to learn these things in public school... I'd be illiterate and have no understanding of mathematics... not enough to program a computer anyway.
The majority of my former school peers couldn't solve equations or even perform long division by their senior year... skills I acquired easily and early on. The schools I attended were award winning, well funded, and regarded as some of the best... and still basic math and reading could not be taught. Nothing new was introduced between the first grade and tenth grade.
Hard sciences are declining because American education is a failure. Our embarrasing literacy rates, highscool drop out rates, and falling standardized test scores is all the proof anyone needs. Kids can't program anymore. We don't give them the skills nor the education.
National ID cards promoted by a competant government are a great idea. It's a matter of trust and I don't trust my government to make informed and responsible decisions regarding the wellfare of its people. I don't trust that they can ever streamline or make much of anything efficient. The DMV, IRS, and public education come to mind. Even the voting machines they contract Diebold to build can't do basic counting correctly. Does anybody really trust them to handle the personal information of everybody?
No exaggeration: Today, you can purchase private information on anybody... courtesy of the DMV, law enforcement, and gross mismanagement by federal agencies. If it's not sold, it's handled so sloppily that it's practically free. I'd much rather these practices stopped now, instead of the empty promises that it won't persist with new cards... as if the physical ID is the problem and not the incompetance of these agencies.
They've thrown everything they can at persuading the public to accept ID cards. It'll help anti-terrorism efforts and somehow track sex offenders. It'll prevent identity theft and end illegal imigration. And don't forget... it's patriotic. I'm sure the twin towers will be printed somewhere on the card... as if that makes the whole thing a good idea.
There are no good justifications in my opinion... not until a rational long-term plan is proposed that takes ALL issues under consideration.
Do we really want the same mandated personal identity requirements that the Russians had under Stalin?
Your points are valid. I hope I didn't come off as a jerk or know it all in my posting. Let me elaborate:
I've built a handful of game engines and I can honestly say... code reuse is a myth in *my* projects. Sure... if you're developing first person shooters or games consistent with an established theme (like an RPG you've already written), then sure... code reuse is a definate possibility and likelyhood... especially among sequels.
When you start pushing the edge of what hardware can do, often these changes affect the entire system. Upgrading your scenegraph to use BSP's requires massive rewrites of all your environment rendering code. There's no real escape from that. New lighting processing will affect every render node. Particle system changes also have the potential to affect an entire physics simulation system.
The interconnected nature is unlike many software systems. This is why I consider it unique and special. Good management and planning is essential, but often ditched to conform to a release date (in my experience).
If I were say... writing an accounting system. Changing the underlying RDBMS or the ledger balance routines rarely affects the ENTIRE system. Good layering can isolate the effect of such changes. Software, not concerned with attaining the highest levels of performance, tends to be much more modular. One of the first steps of optimization (for me) is to remove layers and unnecessary abstraction.
Games often call for engineers to unroll loops, optimize for processor cache hits, and assembler (vectorization). 9/10ths of developers rarely need worry about this gruntwork. It makes for unmaintainable and unreusable code after it's been optimized. You are basically building an implementation specific throw-away modules.
This is why I think it's different... not to engineers accustomed to high performance software... to managers that think all software is the same. It's simply not. The disciplines must match the industry you're targeting.
Evangelizing and support should be left to companies like Red Hat, Novell, and every other company who's business model is built on providing such services. You CANNOT write 10,000's lines of code, document, and support by yourself. Open source developers are not supermen... that's why they seek community support.
What I was trying to say about the documentation:
We (open source developers) spend hours on documentation. There's technical docs for administrators and system integrators. There are readme files and FAQ's for the novice and advanced users alike. There are design documents for coders and contributers. There's source code documentation. That's more than most people have typed in their entire life.
When you put that much effort into pointing people in the right direction you expect that they'll take advantage of that work. When users say "I know there are docs, but..." it's like they're saying "I don't want to read the docs. Stop what you're doing and send me the answer so I don't have to look for it." That's offensive.
Opensource projects often dump their entire body of knowledge into an easyly searchable repository intended for all people. The file that is named "README" often tells you where to go next... as does any greeting dialog box or command-line help.
What I object to is the pretentious users that expect you to not only give them the answers, but to search for them as well. Everyone's time is valuable, and the lazy novice's demands burden everyone. We have things more serious to do than type searches. How many newbies ever try Google?
I'm not one an elitest open source coder. I do believe that software should be accessible to all interested parties. That's why I spend an incredible amount of time testing and designing for usability. I spend dozens of hours documenting everything I develop. It's all right there, for whomever wants to look... and I'll point the lost users to the right place... but I'll never hold someone's hand because they are being lazy.
I work hard to write software. I demand that my users work just a LITTLE to understand it. It's only fair. I do my best to keep the learning curve low. If it's not good enough, I'm here to help. That's fair in my mind. If my docs bite, PLEASE email me and ask questions... however... if the docs are excellent and easy to understand and the user hasn't read them... tough. I'll suggest they read them first. If they get militant, as many have done, I'll ignore them. I'm not obligated to donate every free hour of my time for people that won't help me help them. That's time away from my family and kids... the very reason my first line of support is in the form of easily readable documentation.
There are support groups and forums for entry level users. It's not like I'm saying leave people out to dry. I just can't support every person and rarely do I need to. It's just the occassional ungrateful jerk who's upset I won't go to his house and show him how everything works. It's not enough I'm giving away software I can easily retire by selling.
Remember Classic mode? Sure, it's pokey, but consider it a rough draft. Imagine a classic mode for WinXP. With the proper drivers, you could virtualize the entire operating system, boot it, and share the same screen realestate. The end result: Seemless desktop integration. Perhaps they could even use RDP natively and skip custom drivers.
Their multiple API layers is very impressive. The BSD, Carbon, Cocoa, and Java API's are all very different ways to access the same operating system. If they did add a Win32 layer, which is completely possible, they could (in theory) run every Windows application ever produced... and with less bugs. I had this idea last week. Personally, I believe that this must be a long-term strategy... but I'm not holding my breath.
The problem with speculation like this, however, is the great potential for dissapointment. Running XP virtualized in a window is good enough for me. Window manager integration is even better. API level compatibility is the holy grail, but unlikely... even for Apple. It's almost like anything they do now, might come up short of expectations.
OS/2's Windows 3.1 compatibility comes to mind. It was released right before the shift to Win32 applications (Win32 *S* didn't count as real 32-bit). OS/2 didn't add Win95 compatibility and a whole new generation of apps weren't compatible.
Supporting Win32 may not be enough. Like OS/2, Mac OS is getting Windows support (even through dual-booting) right before a brand-new Windows release and API. If Leopard had Win32 *AND* Vista support... well... that would be enough.
I just hope that Windows compatibility doesn't cause developers to scrap native Mac ports of software. Mac "support" becomes a thing of the past.
I don't believe that games are super artistic, but only because most kids don't really like art that much. I wasn't exactly dying to hang out at the art museum when I was younger. I don't believe it's necessarily a dislike for art as much as it's a preference for things more fun.
So are video games and so is art. People keep producing what other people keep buying. As long as the most valuable market segment never matures, then the games never have to go far beyond what they are right now. The target audience is perpetually sixteen years old. Games don't try to match the maturity of the player and they don't have to grow up with the player. They don't even have to be original.
I guess what I'm trying to say is: It's not the games fault that they're not expansive artistic expressions of infinitely interactive universe. Kids don't necessarily want that. It's all about instant gratification, and that's not very artistic at all.
It's not like newbies don't have Linux communities that cater to them.
It's insulting and arrogant for a newbie to come along and assume that THEY don't need to read the documentation, when they obviously don't understand what's going on. What makes them so special that they deserve personalized support? If hand holding is what you need, then buy Red Hat Linux.
Red Hat, and others, offer support for Linux. You can't demand corporate funded quality of service with absolutely no funding. This is why they survive. They fill a need beyond packaging distros.
Open source has been infiltrated by the "freebie" minded. Downloading code doesn't entitle you to the undying support of everyone involved. Open source developers don't owe anyone. They give away their code to exploit to your own desire. Don't get upset that we won't sit down and explain every detail to you also. We do our best writing docs... something we are also not obligated to do.
Give a little and the "freebie" minded demand more. This type of thinking will kill opensource. It takes little more than a group of lazy ingrates to demotivate a community.
Open source software projects can rarely afford to run a helpdesk. To benefit from free software, you must learn how it works on your own. If you're having a serious inexplicable problem, most developers are glad to help. Everyone accepts this except the newbie, who often feels entitled to free software and free hand holding.
What other industry caters to the lowest-level of human? Does anyone really expect Ford or Chevrolet to answer questions about where the key goes? No. It's in the manual that comes with the car. Does Tivo walk all of its users through recording a show? No, it's in the manual.
Users of software should be obligated to at least read their manuals, even if they don't understand them. The readme is the first line of defense against newbies. An incredible amount of time is spent answering every conceivable question in a FAQ so it doesn't need to be done everytime they come up.
Users shouldn't feel entitled to free help given that time is DONATED. It takes some arrogance to criticize a charitable developer for not finding the already published answer for them. If this vital time is wasted on simple questions, it pollutes knowledge bases and makes difficult answers harder to find. Newbie questions are bad for the community.
I can say that I've been on the receiving end of the most assinine questions. Most of the answers involve common sense or can be had by reading the directions. I've worked helpdesk, which is why I'll never do it again.
In case you think I'm taking too hard of an edge, here's two real life helpdesk questions (from back in the day - same user):
"My computer is broken. I don't know what happened. It just broke. Please hurry and fix it!" - Turned on the monitor
"There's an error message on my screen. I don't know what it says. Can you come and fix it?"
"What does it say?"
"I don't know!"
- Message reads: "Your printer is out of paper. Insert paper into tray 1"
Opensourcing Mac OS X is the number one way to make the operating system resemble Linux. Some people don't see this as bad, but let me explain. I love Linux.
Mac OS X is much more than the Kernel and UNIX command-line and X11. It's a substantial part, that already is opensourced as Darwin. The real value in Mac OS for developers is the incredibly elegant framework built on very high-level components.
Core Data, Core Image, Core Video, Applescript, XCode, QuickTime, Speech, Finder, Aqua, Quartz Extreme, Cocoa Bundles & NIBS... There's a lot to this O/S and it's not something you can just "open" at any time.
There are innumerable software license restrictions in the video CODECS for QuickTime alone. Display PDF? Unless Adobe wants to open source PDF, that just won't happen. This is one of the nicest features of Mac OS. Fonts and vectors actually render as they'll print. Mac users take it for granted. Windows gets this feature in 2007, but no one's asking Microsoft to opensource Windows or ship a stripped down "free" version.
MacOS stands apart, in part, due to its bullish resistance to what everyone else is doing. Opening the code invites pressure to conform, the absolute worst thing that can happen to this OS. Apple has always been an innovator and is often ahead of the rest of the industry.
I fear that an open source community would pressure Apple to abandon the very things that make the OS unique and cutting edge. Their proprietary solutions make for great software.
I can tell you:
As an Objective C (Cocoa) developer, the memory management woes of C++ are long gone. Network communications are so simple, I feel dumb for ever using sockets. Message delegation is a feature so powerfully simple, it allows me to write a fraction of code for the same functionality.
To use Mac OS effectively, you really do need to "think different." The Frameworks make extensive use of generics and design patterns... something Microsoft has only started to embrace in their new toolsets.
When I look at Linux, it's not even close. It's not an end-user OS and never will be without the very things that makes Mac OS what it is. Linux lacks a decent GUI and productivity tools... even the support of commercial development as a whole.
I don't dislike Linux at all. I use Linux and/or BSD for almost everything... embedded hardware, servers, and even light day-to-day tasks. It's just very raw and continues to be a tad hardcore.
Linux is largely C-based. The talent, Dvorak suggests should be tapped, is composed of mostly C developers. The OS is built completely different than one that uses C++ or Objective C as its primary language.
What makes Dvoraks' comments silly, is that there simply isn't a community of good object oriented developers in open source that aren't already actively working on very important projects. I would much rather that some of them finish Eclipse, instead of helping Apple. I think Apple has a handle on it already.
Dvorak is a militant Windows user. He pokes and prods the Apple community. His predictions are meant only to antagonize Apple users.
This is a common problem that plagues every booming industry... especially advertising. Your bean counters arrive, and apply their "insight" and "wisdom" to running the business and increasing productivity. The end result is deadlines, avoidance of solutions that are too difficult to schedule (or understand), reuse of code and concepts that should be trashed... in all: a bad work environment. Game developers, like myself, strive for the cutting edge. The idea of mandated shortcuts pisses us off.
Game development is a creative art. You can't rush or schedule that kind of a process. No project management book or body of knowledge can overcome this. As long as game publishers drive for more efficiency and output, they will burn out their staff. Game development is a business that needs a bit of fat (free time). You need more freedom to develop and burn code to test new concepts. Investing in throw-away code is almost always a business "no no."
Business folks expect that all problems in computer gaming have known solutions. This idea is false. There's a ton of R&D for just about every algorithm. There's not necessarily a "one size fits all" solution to any given problem. And even a solid algorithm can often be implemented in over a dozen different ways.
I've worked for a couple of places that tried to run game development like regular software engineering projects. They did not succeed. Sometimes, entire industries need to ditch the MBAs and embrace what got them to where they are in the first place. Operating efficiency is only a good thing, so long as it doesn't negatively impact your staff, quality, and sales.
Building games is completely different that any other kind of software development. It needs to managed that way... special needs in mind.
How many of us know gamers that neglect their families and health just to get a few hours of good gaming in? Growing up and out of games doesn't make sense to me. There are simply too many addicted adults. What is it then?
As more of the video game industry becomes infiltrated by the movie studios (Chun Yow Fat is starring in a game), the plots and gameplay will be more influenced by the movie industry... known for precanning cliche plotlines and bending to irrational fears about obscenity. You can bet that Anacondas the game will be just as lame as Anacondas the movie, if the same studio is producing it.
This isn't a problem with reaching the audience or demographics. Games, for the most part, are boring and uninventive. I remember years of ripoff side-scrollers with various movie themes during the 80's and 90's. None of these stands out for any reason at all. They were money grabs. Same game, different theme. It's part of the reason the entire industry crashed in the 80's (although, economic depression did help).
When something groundbreaking does get released, it makes news and people buy. No one's losing interest in games any more than they're losing interest in music. Perhaps people are tired of paying $50 (average new game price) for garbage. The price and quality of music certainly caused the masses to "lose interest."
Because the gaming industry is such a booming industry, you're going to get more companies producing games. Sadly, the majority of these will be bad and uninspired. Youth demands that games be interesting, entertaining, and not condescending. Can the industry really expect to keep the attention of the masses after the 28,392nd first person shooter?
Wal-Mart's policies also bear a large influence on movies and music. Artists that refuse to release a "clean" version of a CD have long been banned in retail chains like Wal-Mart and even Target.
Video game manufacturers will simply have to deal with it. I wouldn't want to sell potentially obscene material at my store either (if I owned one). There are plenty of others that will.
When will the media industries grow some balls and produce what they want? If Wal-Mart stops carrying most titles, people will learn to shop elsewhere. If only American business was daring enough.
I like how Dell's resistance to change isn't noted. They didn't exactly embrace USB or IEEE-1394 (FireWire). They took their time offering DVI connectors on desktops. They weren't even at the forefront of SATA either. So we should give them credit for these three "contributions"?
Apple was one of the more active members of the WI-FI consortium. I'm almost positive that they equipped all of their machines with Airport slots and integrated antennas... long before Dell even offered it as an option (by many months). I think even e-Machines beat Dell to the WI-FI race.
The industry had already agreed on 802.11 long before Intel decided to follow suit (Intel's horrid Home-RF). I'm highly skeptical of Dell claiming that their clout had anything to do with brining Intel to their senses. Not even Intel can take on the whole industry... just look at Itanium.
So they may have helped convince Intel to bring 64-bit to the Pentium like the AMD Opteron, PowerPC, MIPS, SPARC, PA-RISC, Itanium, and Alpha. Who cares? It was a natural evolution that software (on PCs) still can barely take advantage of, barring massive recompilation. Everybody knows that AMD, is the indisputable innovator and influence for x86-64... beating Intel out the door by a large margin.
Dell backed 64-bit extensions to x86, primarily based on price. Itanium was simply out of the price point for their customers. No one should receive credit or be rewarded for running away from Itanium. Especially when your biggest reason is price. There's 1,000,001 reasons not to use Itanium. I'd be more impressed if they actually named one.
Lastly, the article makes it sound as if Dell invented PCI Express in-house. Dell doesn't design chipsets. They only buy them in bulk quantity. It's as if I'd take credit for the rise of Jolt Cola, because I buy it in bulk at Costco. No no no. They tell their fab to solder whatever connectors the chipsets support. It's not like they led the drive for faster interconnects. Again, AMD led the pack with Hypertransport, borrowed from the Alpha team... pushing Intel to raise their FSB speed, to even be able to support PCI-X & PCI-E.
I wasn't trying to infer that anybody has "stolen" code, however, I think that giving back includes more than a short disclaimer... especially for profitable enterprises.
It's in everybody's best interest to let these engineers continue to develop as they have been. More than just OpenSSH & OpenSSL has come out of BSD projects. Let's make sure that continues to happen by giving the projects their deserved recognition.
I've noticed some undue emphasis placed on OpenSSH & OpenSSL. They are GREAT packages, but not the only thing people benefit from. Don't forget, that nearly every commercial operating system has pilfered code from the BSD projects.
EVERYBODY should contribute, especially the companies that have profited from the hard work of the team.
This book was Slashdotted a few months ago. It's written and endorsed by members of the core PHP team. It's the most accurate language resource I've come across. It covers the PHP language, while I've found that other books tend to offer cookie-cutter "recipes" for common scenarios (code snippets). Sadly, its mention of interfaces is a bit sparse and it pre-dates PHP 5.1's PDO. Nevertheless, I still find it very relevant for anyone doing OOP with PHP. There are one or two gems not found in the online documentation.
NOTE: It's better to have some PHP programming experience before reading.
I was a big fan... until I needed to use PostgreSQL 7 for a real (commercially available) product. To call it slow would be an abomination of the word. Slow doesn't even begin to describe b-tree insert times. Yes, I tuned the engine and dropped indexes at the appropriate times. Yes, my data structures were relational. Yes, this contradicts some published benchmarks. My use is real world and in reality, PostgreSQL is slow... and a bit buggy.
Nested parentheses in SQL can cause an engine crash. " like... (SELECT A INNER JOIN B) INNER JOIN..." But the crashing is tolerable. Hand-holding the query optimizer is not. Quite often, the optimizer gets the query plan wrong. Sending special commands to disable internal features is often the only resort.
While it's true that PostgreSQL is more database than most corporate weenies need, it falls down in moderate write environments. It's best used for systems that write data very infrequently, otherwise it fragments quickly. The only solution to table and database fragmentation is dump & reload.
Vacuum is asinine. Any command that needs to be run periodically under threat of complete and total data corruption should not be. That's right. Only PostgreSQL makes you vacuum or else your transaction ids overflow. This is modern? I'm shocked.
Literacy rates in the early part of the century, up until the 1950's, were high compared to today. In recent years, we have been seeing a sharp decline. Comic books were one of the first scapegoats, along with dancing, masturbation, radio, television, and color of skin. The numbers are still falling and the newest scapegoat is... IM's and email. *belCh*
I actually read that pagers were to blame in one mid-90's article. The act of spelling with a touchtone keypad was somehow dumbing kids down. Cell phones, video games, lack of after school programs, television, cartoons, violence, fast food, candy... We've all heard at least one of these items implicated as a contributor to illiteracy. I believe the cause or at least, the most significant contributor, is education.
A few decades ago, our educators decided to undertake an experiment with language and reading. They eliminated the use of phonics and replaced it with a new repetition based training method (Cat in the Hat is an example). Instead of constructing words by sound, children were encouraged to learn spelling and reading by vocabulary drills and other tasks concerned with memorization alone. Repetition was the key to effective teaching in this new philosophy.
When some children began to have problems with the new curriculum, educators didn't necessarily view it as failure. Some of these students were put into categories like "special needs." People often wrongly assume that public education is a scientific institution. They reflect the full range of human biases for the period. They aren't bound by the scientific method, so they do not use it.
It would be unfair and shitty to make Apple allow Mac OS X on all Intel-ish hardware. They would have to write drivers for ALL of the Taiwanese garbage equipment out there? c'mon and be realistic. That's not a good or desirable idea by anyone. Windows doesn't even work smoothly on some of these new clone chip-sets.
Don't give me some speech about how this clone PC is just as good, but costs less crap. That's not the point. Standardization of drivers, quality control, and management of the platform is what gives it extraordinary benefit to developers. It keeps Apple's costs down and allows them to add features to the O/S and not just bolster support. Microsoft has been caught up in the support nightmare. They'd lock down the PC platform in an instant... if they could.
I know everybody with a $499 Dell PC wants to run OS X. I want a LS2 engine inside my shitty Ford. Neither are good ideas from an engineering perspective. The shorter PC manufacturing cycle introduces errata that MUST be compensated for. Apple has it too. Manufacturers can't update the operating system and support libraries and keep up with the rest of the industry! There's something to be said about only supporting yourself.
There are niceties I'm not willing to trade so that others can run OS X. I'd *HATE* to need to assume that vectorization support might be missing from some Intel Macs, because SOME of them use AMD or even VIA processors. FUCK THAT.
This isn't about race, but the x86 isn't pure. I think of it as a collection of pseudo-standards that play well together. Intel code is neither optimized nor completely portable to AMD CPU, and vice-versa. Memory controllers are varied. Even the IDE controllers can be different... something as standard as an IDE controller!
I hope Apple never opens it up. That is the day that innovation dies and support begins.
First, the uninstaller is broken. Plan on reformatting before you play with this beast. Also, plan on 2GB+ of RAM. The 1GB minimum is a lie even for a single user. Also, if your source repository is larger than 4GB, you'll need SQL Server 2005. It's demanding on the clients also. Plan on generous RAM & CPU speed.
I can't migrate my company's Visual SourceSafe repository. The built-in analyze tools to repair corruption lock up before they do anything useful. VSS corrupts by nature. Since my repository is trashed and the repair tools are broken, I have no way of importing my source tree... into the less-mature and unreleased (still beta) Team Server.
Keep in mind, TFS's source code management is supposed to be better than VSS... not a replacement for Subversion. That's a big difference to keep note of. My most recent gripe: Rollbacks are sometimes impossible for merges. There is also no help documentation on this product, aside from marketing fluff. Even the docs are in beta.
If Team Foundation Server 1.0 is anything like Visual SourceSafe 1.0, keep your eyes open.
Slashdot Mirror
Call the SBC's DSL department and claim to be a friend "helping" someone install their DSL modem... but insist that you don't know the address or anything else. Be as dumb as possible on the phone. Get a little drunk if you can't be convincing.
Often, the customer service reps will read back the entire address, and sometimes, even the last for digits of the SSN. I found this out when I was ligitimately calling them because of a line problem.
I never had any problems adding service, removing service, or getting personal account information... all without identifying myself whatsoever. Need an address for a telephone number, call SBC and tell them you want DSL. The phone reps will "verify" your address by reading it back. Awesome, huh?
"...there is only a small chance his brother will get in."
Let's hope not. I'm pretty sure the Bush family bankrupted Texas in governorship.
The ZCE hasn't earned even one extra dollar for me, but that's not why I have it. I needed wall art for my home office.
Certification is rather useless for ascertaining skill level, but does filter out the lowest classes of developer. I use certifications to skip really basic interview questions... for that, the ZCE is super useful. If you interview a lot, it's worth investing in.
Some of the noisiest network traffic is caused by BitTorrent. It's responsible for so many false alarms that organizations tend to ban it completely.
I'm sure ISPs will hate this and/or forbid it... so I must support it.
My mother taught me how to read and my father taught me basic algebra. I started programming at age seven, already having the necessary skills. If I was forced to learn these things in public school... I'd be illiterate and have no understanding of mathematics... not enough to program a computer anyway.
The majority of my former school peers couldn't solve equations or even perform long division by their senior year... skills I acquired easily and early on. The schools I attended were award winning, well funded, and regarded as some of the best... and still basic math and reading could not be taught. Nothing new was introduced between the first grade and tenth grade.
Hard sciences are declining because American education is a failure. Our embarrasing literacy rates, highscool drop out rates, and falling standardized test scores is all the proof anyone needs. Kids can't program anymore. We don't give them the skills nor the education.
National ID cards promoted by a competant government are a great idea. It's a matter of trust and I don't trust my government to make informed and responsible decisions regarding the wellfare of its people. I don't trust that they can ever streamline or make much of anything efficient. The DMV, IRS, and public education come to mind. Even the voting machines they contract Diebold to build can't do basic counting correctly. Does anybody really trust them to handle the personal information of everybody?
No exaggeration: Today, you can purchase private information on anybody... courtesy of the DMV, law enforcement, and gross mismanagement by federal agencies. If it's not sold, it's handled so sloppily that it's practically free. I'd much rather these practices stopped now, instead of the empty promises that it won't persist with new cards... as if the physical ID is the problem and not the incompetance of these agencies.
They've thrown everything they can at persuading the public to accept ID cards. It'll help anti-terrorism efforts and somehow track sex offenders. It'll prevent identity theft and end illegal imigration. And don't forget... it's patriotic. I'm sure the twin towers will be printed somewhere on the card... as if that makes the whole thing a good idea.
There are no good justifications in my opinion... not until a rational long-term plan is proposed that takes ALL issues under consideration.
Do we really want the same mandated personal identity requirements that the Russians had under Stalin?
Your points are valid. I hope I didn't come off as a jerk or know it all in my posting. Let me elaborate:
I've built a handful of game engines and I can honestly say... code reuse is a myth in *my* projects. Sure... if you're developing first person shooters or games consistent with an established theme (like an RPG you've already written), then sure... code reuse is a definate possibility and likelyhood... especially among sequels.
When you start pushing the edge of what hardware can do, often these changes affect the entire system. Upgrading your scenegraph to use BSP's requires massive rewrites of all your environment rendering code. There's no real escape from that. New lighting processing will affect every render node. Particle system changes also have the potential to affect an entire physics simulation system.
The interconnected nature is unlike many software systems. This is why I consider it unique and special. Good management and planning is essential, but often ditched to conform to a release date (in my experience).
If I were say... writing an accounting system. Changing the underlying RDBMS or the ledger balance routines rarely affects the ENTIRE system. Good layering can isolate the effect of such changes. Software, not concerned with attaining the highest levels of performance, tends to be much more modular. One of the first steps of optimization (for me) is to remove layers and unnecessary abstraction.
Games often call for engineers to unroll loops, optimize for processor cache hits, and assembler (vectorization). 9/10ths of developers rarely need worry about this gruntwork. It makes for unmaintainable and unreusable code after it's been optimized. You are basically building an implementation specific throw-away modules.
This is why I think it's different... not to engineers accustomed to high performance software... to managers that think all software is the same. It's simply not. The disciplines must match the industry you're targeting.
Evangelizing and support should be left to companies like Red Hat, Novell, and every other company who's business model is built on providing such services. You CANNOT write 10,000's lines of code, document, and support by yourself. Open source developers are not supermen... that's why they seek community support.
What I was trying to say about the documentation:
We (open source developers) spend hours on documentation. There's technical docs for administrators and system integrators. There are readme files and FAQ's for the novice and advanced users alike. There are design documents for coders and contributers. There's source code documentation. That's more than most people have typed in their entire life.
When you put that much effort into pointing people in the right direction you expect that they'll take advantage of that work. When users say "I know there are docs, but..." it's like they're saying "I don't want to read the docs. Stop what you're doing and send me the answer so I don't have to look for it." That's offensive.
Opensource projects often dump their entire body of knowledge into an easyly searchable repository intended for all people. The file that is named "README" often tells you where to go next... as does any greeting dialog box or command-line help.
What I object to is the pretentious users that expect you to not only give them the answers, but to search for them as well. Everyone's time is valuable, and the lazy novice's demands burden everyone. We have things more serious to do than type searches. How many newbies ever try Google?
I'm not one an elitest open source coder. I do believe that software should be accessible to all interested parties. That's why I spend an incredible amount of time testing and designing for usability. I spend dozens of hours documenting everything I develop. It's all right there, for whomever wants to look... and I'll point the lost users to the right place... but I'll never hold someone's hand because they are being lazy.
I work hard to write software. I demand that my users work just a LITTLE to understand it. It's only fair. I do my best to keep the learning curve low. If it's not good enough, I'm here to help. That's fair in my mind. If my docs bite, PLEASE email me and ask questions... however... if the docs are excellent and easy to understand and the user hasn't read them... tough. I'll suggest they read them first. If they get militant, as many have done, I'll ignore them. I'm not obligated to donate every free hour of my time for people that won't help me help them. That's time away from my family and kids... the very reason my first line of support is in the form of easily readable documentation.
There are support groups and forums for entry level users. It's not like I'm saying leave people out to dry. I just can't support every person and rarely do I need to. It's just the occassional ungrateful jerk who's upset I won't go to his house and show him how everything works. It's not enough I'm giving away software I can easily retire by selling.
Remember Classic mode? Sure, it's pokey, but consider it a rough draft. Imagine a classic mode for WinXP. With the proper drivers, you could virtualize the entire operating system, boot it, and share the same screen realestate. The end result: Seemless desktop integration. Perhaps they could even use RDP natively and skip custom drivers.
Their multiple API layers is very impressive. The BSD, Carbon, Cocoa, and Java API's are all very different ways to access the same operating system. If they did add a Win32 layer, which is completely possible, they could (in theory) run every Windows application ever produced... and with less bugs. I had this idea last week. Personally, I believe that this must be a long-term strategy... but I'm not holding my breath.
The problem with speculation like this, however, is the great potential for dissapointment. Running XP virtualized in a window is good enough for me. Window manager integration is even better. API level compatibility is the holy grail, but unlikely... even for Apple. It's almost like anything they do now, might come up short of expectations.
OS/2's Windows 3.1 compatibility comes to mind. It was released right before the shift to Win32 applications (Win32 *S* didn't count as real 32-bit). OS/2 didn't add Win95 compatibility and a whole new generation of apps weren't compatible.
Supporting Win32 may not be enough. Like OS/2, Mac OS is getting Windows support (even through dual-booting) right before a brand-new Windows release and API. If Leopard had Win32 *AND* Vista support... well... that would be enough.
I just hope that Windows compatibility doesn't cause developers to scrap native Mac ports of software. Mac "support" becomes a thing of the past.
I don't believe that games are super artistic, but only because most kids don't really like art that much. I wasn't exactly dying to hang out at the art museum when I was younger. I don't believe it's necessarily a dislike for art as much as it's a preference for things more fun.
So are video games and so is art. People keep producing what other people keep buying. As long as the most valuable market segment never matures, then the games never have to go far beyond what they are right now. The target audience is perpetually sixteen years old. Games don't try to match the maturity of the player and they don't have to grow up with the player. They don't even have to be original.
I guess what I'm trying to say is: It's not the games fault that they're not expansive artistic expressions of infinitely interactive universe. Kids don't necessarily want that. It's all about instant gratification, and that's not very artistic at all.
It's not like newbies don't have Linux communities that cater to them.
It's insulting and arrogant for a newbie to come along and assume that THEY don't need to read the documentation, when they obviously don't understand what's going on. What makes them so special that they deserve personalized support? If hand holding is what you need, then buy Red Hat Linux.
Red Hat, and others, offer support for Linux. You can't demand corporate funded quality of service with absolutely no funding. This is why they survive. They fill a need beyond packaging distros.
Open source has been infiltrated by the "freebie" minded. Downloading code doesn't entitle you to the undying support of everyone involved. Open source developers don't owe anyone. They give away their code to exploit to your own desire. Don't get upset that we won't sit down and explain every detail to you also. We do our best writing docs... something we are also not obligated to do.
Give a little and the "freebie" minded demand more. This type of thinking will kill opensource. It takes little more than a group of lazy ingrates to demotivate a community.
Open source software projects can rarely afford to run a helpdesk. To benefit from free software, you must learn how it works on your own. If you're having a serious inexplicable problem, most developers are glad to help. Everyone accepts this except the newbie, who often feels entitled to free software and free hand holding.
What other industry caters to the lowest-level of human? Does anyone really expect Ford or Chevrolet to answer questions about where the key goes? No. It's in the manual that comes with the car. Does Tivo walk all of its users through recording a show? No, it's in the manual.
Users of software should be obligated to at least read their manuals, even if they don't understand them. The readme is the first line of defense against newbies. An incredible amount of time is spent answering every conceivable question in a FAQ so it doesn't need to be done everytime they come up.
Users shouldn't feel entitled to free help given that time is DONATED. It takes some arrogance to criticize a charitable developer for not finding the already published answer for them. If this vital time is wasted on simple questions, it pollutes knowledge bases and makes difficult answers harder to find. Newbie questions are bad for the community.
I can say that I've been on the receiving end of the most assinine questions. Most of the answers involve common sense or can be had by reading the directions. I've worked helpdesk, which is why I'll never do it again.
In case you think I'm taking too hard of an edge, here's two real life helpdesk questions (from back in the day - same user):
"My computer is broken. I don't know what happened. It just broke. Please hurry and fix it!"
- Turned on the monitor
"There's an error message on my screen. I don't know what it says. Can you come and fix it?"
"What does it say?"
"I don't know!"
- Message reads: "Your printer is out of paper. Insert paper into tray 1"
Opensourcing Mac OS X is the number one way to make the operating system resemble Linux. Some people don't see this as bad, but let me explain. I love Linux.
Mac OS X is much more than the Kernel and UNIX command-line and X11. It's a substantial part, that already is opensourced as Darwin. The real value in Mac OS for developers is the incredibly elegant framework built on very high-level components.
Core Data, Core Image, Core Video, Applescript, XCode, QuickTime, Speech, Finder, Aqua, Quartz Extreme, Cocoa Bundles & NIBS... There's a lot to this O/S and it's not something you can just "open" at any time.
There are innumerable software license restrictions in the video CODECS for QuickTime alone. Display PDF? Unless Adobe wants to open source PDF, that just won't happen. This is one of the nicest features of Mac OS. Fonts and vectors actually render as they'll print. Mac users take it for granted. Windows gets this feature in 2007, but no one's asking Microsoft to opensource Windows or ship a stripped down "free" version.
MacOS stands apart, in part, due to its bullish resistance to what everyone else is doing. Opening the code invites pressure to conform, the absolute worst thing that can happen to this OS. Apple has always been an innovator and is often ahead of the rest of the industry.
I fear that an open source community would pressure Apple to abandon the very things that make the OS unique and cutting edge. Their proprietary solutions make for great software.
I can tell you:
As an Objective C (Cocoa) developer, the memory management woes of C++ are long gone. Network communications are so simple, I feel dumb for ever using sockets. Message delegation is a feature so powerfully simple, it allows me to write a fraction of code for the same functionality.
To use Mac OS effectively, you really do need to "think different." The Frameworks make extensive use of generics and design patterns... something Microsoft has only started to embrace in their new toolsets.
When I look at Linux, it's not even close. It's not an end-user OS and never will be without the very things that makes Mac OS what it is. Linux lacks a decent GUI and productivity tools... even the support of commercial development as a whole.
I don't dislike Linux at all. I use Linux and/or BSD for almost everything... embedded hardware, servers, and even light day-to-day tasks. It's just very raw and continues to be a tad hardcore.
Linux is largely C-based. The talent, Dvorak suggests should be tapped, is composed of mostly C developers. The OS is built completely different than one that uses C++ or Objective C as its primary language.
What makes Dvoraks' comments silly, is that there simply isn't a community of good object oriented developers in open source that aren't already actively working on very important projects. I would much rather that some of them finish Eclipse, instead of helping Apple. I think Apple has a handle on it already.
Dvorak is a militant Windows user. He pokes and prods the Apple community. His predictions are meant only to antagonize Apple users.
This is a common problem that plagues every booming industry... especially advertising. Your bean counters arrive, and apply their "insight" and "wisdom" to running the business and increasing productivity. The end result is deadlines, avoidance of solutions that are too difficult to schedule (or understand), reuse of code and concepts that should be trashed... in all: a bad work environment. Game developers, like myself, strive for the cutting edge. The idea of mandated shortcuts pisses us off.
Game development is a creative art. You can't rush or schedule that kind of a process. No project management book or body of knowledge can overcome this. As long as game publishers drive for more efficiency and output, they will burn out their staff. Game development is a business that needs a bit of fat (free time). You need more freedom to develop and burn code to test new concepts. Investing in throw-away code is almost always a business "no no."
Business folks expect that all problems in computer gaming have known solutions. This idea is false. There's a ton of R&D for just about every algorithm. There's not necessarily a "one size fits all" solution to any given problem. And even a solid algorithm can often be implemented in over a dozen different ways.
I've worked for a couple of places that tried to run game development like regular software engineering projects. They did not succeed. Sometimes, entire industries need to ditch the MBAs and embrace what got them to where they are in the first place. Operating efficiency is only a good thing, so long as it doesn't negatively impact your staff, quality, and sales.
Building games is completely different that any other kind of software development. It needs to managed that way... special needs in mind.
How many of us know gamers that neglect their families and health just to get a few hours of good gaming in? Growing up and out of games doesn't make sense to me. There are simply too many addicted adults. What is it then?
As more of the video game industry becomes infiltrated by the movie studios (Chun Yow Fat is starring in a game), the plots and gameplay will be more influenced by the movie industry... known for precanning cliche plotlines and bending to irrational fears about obscenity. You can bet that Anacondas the game will be just as lame as Anacondas the movie, if the same studio is producing it.
This isn't a problem with reaching the audience or demographics. Games, for the most part, are boring and uninventive. I remember years of ripoff side-scrollers with various movie themes during the 80's and 90's. None of these stands out for any reason at all. They were money grabs. Same game, different theme. It's part of the reason the entire industry crashed in the 80's (although, economic depression did help).
When something groundbreaking does get released, it makes news and people buy. No one's losing interest in games any more than they're losing interest in music. Perhaps people are tired of paying $50 (average new game price) for garbage. The price and quality of music certainly caused the masses to "lose interest."
Because the gaming industry is such a booming industry, you're going to get more companies producing games. Sadly, the majority of these will be bad and uninspired. Youth demands that games be interesting, entertaining, and not condescending. Can the industry really expect to keep the attention of the masses after the 28,392nd first person shooter?
Wal-Mart's policies also bear a large influence on movies and music. Artists that refuse to release a "clean" version of a CD have long been banned in retail chains like Wal-Mart and even Target.
Video game manufacturers will simply have to deal with it. I wouldn't want to sell potentially obscene material at my store either (if I owned one). There are plenty of others that will.
When will the media industries grow some balls and produce what they want? If Wal-Mart stops carrying most titles, people will learn to shop elsewhere. If only American business was daring enough.
I like how Dell's resistance to change isn't noted. They didn't exactly embrace USB or IEEE-1394 (FireWire). They took their time offering DVI connectors on desktops. They weren't even at the forefront of SATA either. So we should give them credit for these three "contributions"?
Apple was one of the more active members of the WI-FI consortium. I'm almost positive that they equipped all of their machines with Airport slots and integrated antennas... long before Dell even offered it as an option (by many months). I think even e-Machines beat Dell to the WI-FI race.
The industry had already agreed on 802.11 long before Intel decided to follow suit (Intel's horrid Home-RF). I'm highly skeptical of Dell claiming that their clout had anything to do with brining Intel to their senses. Not even Intel can take on the whole industry... just look at Itanium.
So they may have helped convince Intel to bring 64-bit to the Pentium like the AMD Opteron, PowerPC, MIPS, SPARC, PA-RISC, Itanium, and Alpha. Who cares? It was a natural evolution that software (on PCs) still can barely take advantage of, barring massive recompilation. Everybody knows that AMD, is the indisputable innovator and influence for x86-64... beating Intel out the door by a large margin.
Dell backed 64-bit extensions to x86, primarily based on price. Itanium was simply out of the price point for their customers. No one should receive credit or be rewarded for running away from Itanium. Especially when your biggest reason is price. There's 1,000,001 reasons not to use Itanium. I'd be more impressed if they actually named one.
Lastly, the article makes it sound as if Dell invented PCI Express in-house. Dell doesn't design chipsets. They only buy them in bulk quantity. It's as if I'd take credit for the rise of Jolt Cola, because I buy it in bulk at Costco. No no no. They tell their fab to solder whatever connectors the chipsets support. It's not like they led the drive for faster interconnects. Again, AMD led the pack with Hypertransport, borrowed from the Alpha team... pushing Intel to raise their FSB speed, to even be able to support PCI-X & PCI-E.
Everything is all lies.
I wasn't trying to infer that anybody has "stolen" code, however, I think that giving back includes more than a short disclaimer... especially for profitable enterprises.
It's in everybody's best interest to let these engineers continue to develop as they have been. More than just OpenSSH & OpenSSL has come out of BSD projects. Let's make sure that continues to happen by giving the projects their deserved recognition.
The day OpenBSD goes away is a sad one indeed.
I've noticed some undue emphasis placed on OpenSSH & OpenSSL. They are GREAT packages, but not the only thing people benefit from. Don't forget, that nearly every commercial operating system has pilfered code from the BSD projects.
EVERYBODY should contribute, especially the companies that have profited from the hard work of the team.
my bad
This book was Slashdotted a few months ago. It's written and endorsed by members of the core PHP team. It's the most accurate language resource I've come across. It covers the PHP language, while I've found that other books tend to offer cookie-cutter "recipes" for common scenarios (code snippets). Sadly, its mention of interfaces is a bit sparse and it pre-dates PHP 5.1's PDO. Nevertheless, I still find it very relevant for anyone doing OOP with PHP. There are one or two gems not found in the online documentation.
NOTE: It's better to have some PHP programming experience before reading.
ISBN: 0-13-147149-X
I was a big fan... until I needed to use PostgreSQL 7 for a real (commercially available) product. To call it slow would be an abomination of the word. Slow doesn't even begin to describe b-tree insert times. Yes, I tuned the engine and dropped indexes at the appropriate times. Yes, my data structures were relational. Yes, this contradicts some published benchmarks. My use is real world and in reality, PostgreSQL is slow... and a bit buggy.
... (SELECT A INNER JOIN B) INNER JOIN ..." But the crashing is tolerable. Hand-holding the query optimizer is not. Quite often, the optimizer gets the query plan wrong. Sending special commands to disable internal features is often the only resort.
Nested parentheses in SQL can cause an engine crash. " like
While it's true that PostgreSQL is more database than most corporate weenies need, it falls down in moderate write environments. It's best used for systems that write data very infrequently, otherwise it fragments quickly. The only solution to table and database fragmentation is dump & reload.
Vacuum is asinine. Any command that needs to be run periodically under threat of complete and total data corruption should not be. That's right. Only PostgreSQL makes you vacuum or else your transaction ids overflow. This is modern? I'm shocked.
PostgreSQL is to Oracle what Gimp is to Photoshop
Literacy rates in the early part of the century, up until the 1950's, were high compared to today. In recent years, we have been seeing a sharp decline. Comic books were one of the first scapegoats, along with dancing, masturbation, radio, television, and color of skin. The numbers are still falling and the newest scapegoat is ... IM's and email. *belCh*
... We've all heard at least one of these items implicated as a contributor to illiteracy. I believe the cause or at least, the most significant contributor, is education.
I actually read that pagers were to blame in one mid-90's article. The act of spelling with a touchtone keypad was somehow dumbing kids down. Cell phones, video games, lack of after school programs, television, cartoons, violence, fast food, candy
A few decades ago, our educators decided to undertake an experiment with language and reading. They eliminated the use of phonics and replaced it with a new repetition based training method (Cat in the Hat is an example). Instead of constructing words by sound, children were encouraged to learn spelling and reading by vocabulary drills and other tasks concerned with memorization alone. Repetition was the key to effective teaching in this new philosophy.
When some children began to have problems with the new curriculum, educators didn't necessarily view it as failure. Some of these students were put into categories like "special needs." People often wrongly assume that public education is a scientific institution. They reflect the full range of human biases for the period. They aren't bound by the scientific method, so they do not use it.
Who in their right mind wants to support it?
It would be unfair and shitty to make Apple allow Mac OS X on all Intel-ish hardware. They would have to write drivers for ALL of the Taiwanese garbage equipment out there? c'mon and be realistic. That's not a good or desirable idea by anyone. Windows doesn't even work smoothly on some of these new clone chip-sets.
Don't give me some speech about how this clone PC is just as good, but costs less crap. That's not the point. Standardization of drivers, quality control, and management of the platform is what gives it extraordinary benefit to developers. It keeps Apple's costs down and allows them to add features to the O/S and not just bolster support. Microsoft has been caught up in the support nightmare. They'd lock down the PC platform in an instant... if they could.
I know everybody with a $499 Dell PC wants to run OS X. I want a LS2 engine inside my shitty Ford. Neither are good ideas from an engineering perspective. The shorter PC manufacturing cycle introduces errata that MUST be compensated for. Apple has it too. Manufacturers can't update the operating system and support libraries and keep up with the rest of the industry! There's something to be said about only supporting yourself.
There are niceties I'm not willing to trade so that others can run OS X. I'd *HATE* to need to assume that vectorization support might be missing from some Intel Macs, because SOME of them use AMD or even VIA processors. FUCK THAT.
This isn't about race, but the x86 isn't pure. I think of it as a collection of pseudo-standards that play well together. Intel code is neither optimized nor completely portable to AMD CPU, and vice-versa. Memory controllers are varied. Even the IDE controllers can be different... something as standard as an IDE controller!
I hope Apple never opens it up. That is the day that innovation dies and support begins.
First, the uninstaller is broken. Plan on reformatting before you play with this beast. Also, plan on 2GB+ of RAM. The 1GB minimum is a lie even for a single user. Also, if your source repository is larger than 4GB, you'll need SQL Server 2005. It's demanding on the clients also. Plan on generous RAM & CPU speed.
I can't migrate my company's Visual SourceSafe repository. The built-in analyze tools to repair corruption lock up before they do anything useful. VSS corrupts by nature. Since my repository is trashed and the repair tools are broken, I have no way of importing my source tree... into the less-mature and unreleased (still beta) Team Server.
Keep in mind, TFS's source code management is supposed to be better than VSS... not a replacement for Subversion. That's a big difference to keep note of. My most recent gripe: Rollbacks are sometimes impossible for merges. There is also no help documentation on this product, aside from marketing fluff. Even the docs are in beta.
If Team Foundation Server 1.0 is anything like Visual SourceSafe 1.0, keep your eyes open.