Given the tons of garbage on the net, and how half the search topics give you 10,000+ blogs, clan sites, message boards, etc, if you can actually discern between the right info and the bullshit, then IMHO you already knew enough to write the paper on your own.
As a random example, if I were to write a paper about the crusades, based on info on the net, I'd be faced with a deluge of conflicting information. I'd have to discern between posts and pages by people who knew their stuff, and bogus pages and posts by gamers who got their facts from games like Lionheart.
And I doubt that any professor would appreciate if I handed in a paper saying that Richard Coeur de Lion (Lionheart) opened a rift to hell and unleashed demons upon earth. And then allied with Sultan Saladin to fight them off.
Hey, it's in the game Lionheart, so it must be true.
Or: were the crusades started really by Peter the Hermit, to free the holy land, or was it all the Pope's plan to help the Byzantine Empire against the Turks? Would the crusades have still started if the Byzantines weren't starting to take a beating? You can find both points of view presented as historical fact on the net.
Or: which crusade conquered and sacked Constantinople? Was it the fourth, the fifth, or which? And why did that happen? Again, you can find the most conflicting information on the net.
A lot of it is simply conflicting historical data, because different chronicles back then had different biases. But about 75% is simply people taking out of their ass. Just because, you know, it's the Net and you can publish _anything_.
It gets even sadder when it's about programming. Every burger flipper can get hired as one, and God knows every other of them has got to re-publish on the net some "cool tricks" he's found on the net. Half of which don't even work, or are "optimizations" which actually make the program run 10 times _slower_.
So basically it seems to me safer to just read the book and do your own research. By the time you're actually competent to discern the gold from the bullshit on the net, you've already done your research and are more than competent to write your own paper.
There's another reason why your "adding a VCR to an existing TV setup" example is the prime example of what's wrong with computers today.
If you added the VCR wrong, you get feedback. It's obviously not working. You know you need to try again, or get help from someone who knows.
Whereas an insecure WiFi setup gives you no hint at all.
In the quest to _seem_ easy to use, but without actually having to invest in real ease of use, we're just covering up the problems and hoping that noone notices.
Everyone wants their device or program to look like it's just a trivial plug-and-play affair, so what do they do? Maybe actually invest in making it so? Nah, we'll stick to cutting corners instead. We'll make it plug-and-run-wrong (e.g., insecure by default or with a hard-coded admin account) and hope the user doesn't notice it's broken.
And when he does notice that for half a year he's been running a porn server _and_ a spam server _and_ his computer and connection were clogged... we'll just call him an idiot.
In reality, that's simply broken design, not the user's fault.
If TV manufacturers worked that way, instead of giving you digital tuners and a remote control, they'd just default to showing you one preset station so you hopefully don't notice when you've tuned it wrong. (And, hey, just think of the monopoly possibilities and raking the big ad bucks.)
"1. When was the last time someone 0wn3d your TV or VCR?"
Actually, I hope you do realize you've just proven the other guy's point. That computers are such a fragile tool, and for a lot of people they can cause more grief than good, is precisely the _problem_.
What Joe Average wants -- or for that matter what _I_ want -- is something that just does a certain job, with a minimum of fuss. Yes, like a TV or a VCR. If I want to read my email or play a game or whatever, I should just get straight to doing that, instead of having to babysit and secure a piss-poorly made tool.
"2. More complex systems require more complex instructions."
No, it's just a case of letting the idiots run the show. Plain and simple.
Other tools started complicated to use too. Owning a car used to require either being a skilled mechanic yourself, or being rich enough to pretty much hire one full time. Getting an early radio to work, or tune it to a station, was a time-consuming pain in the butt. Etc.
But you know what? Someone in those industries actually cared for the customer. (Or just about the bottom line. Competition is good at that.) Instead of whining about idiot users who can't even learn to use a radio right, they gave you channel presets, auto-scanning for stations, remote controls, and other such.
That's really the only problem with computers today. That instead of asking "how could we make this easier for Joe Average?", we're whining about how Joe is an idot and a luser who can't learn doing things our arcane way.
E.g., if we're talking about wifi, it would be a no-brainer to:
- have a nice wizard interface and walk him through securing the thing.
- make sure that security is enabled by default, and that Joe has to explicitly disable it, if he _really_ wants to run a public "download porn and warez anonymously" service.
- If the device has a default admin username and password, explicitly ask him to change it.
- But what if Joe forgets the password? No problem. Don't fscking have an unchangeable one hardcoded in firmware. Provide an easy way to change it, but which requires physical access to the device. E.g., have to open a lid and press a sunk reset button. After which again, make him change it.
Etc.
See, it didn't even require that much thinking.
But no, instead we'll just whine about how Joe is an idiot luser. Although it's not Joe who's the idiot there.
"3. Adding a wifi router to an existing computer setup is more akin to adding a VCR to an existing TV setup."
I'll direct you to your own point 1: when was the last time someone "0wn3d" your TV after that?
Or if we're talking unneeded complexity, when was the last time you had to become a security expert to add a VCR? Did you have to just know how to generate and share keys on them? And did you need to find that out on your own?
Well, my real curiosity is "why the heck does Canada need _cooling_ the air?"
Look at a world map and see where Hungary is, and where Canada is. Hint: a lot more to the north. It's up there with Siberia.
Or with Scandinavia, but the difference is that Scandinavia has the warm Gulf Stream. Canada doesn't. As I've said, the most apt comparison is Siberia. Another way to put it would be "fucking cold."
So why would anyone need to cool the air is beyond me. Might as well enjoy that warmth while it lasts, IMHO.
Seems to me like what they'd _really_ save energy with, is if they found a way to naturally warm the rooms in winter, rather than on cooling them in what little summer they get.
Then again, I'm sitting at 30 C in an office without air conditioning, and I think it's very nice. So your mileage may vary.
Problem solved only because noone gives a damn about something which has maybe 1/1000 of the market. Mozilla on all platforms is already a spit in the bucket. Mozilla on MacOS, well, that's got to mean... what? You and 10 other users? Hardly a market worth programming for.
The fact is: nothing in MacOS X, Mozilla or even Linux would protect you from spyware.
_Maybe_ I could be persuaded to believe it gives you some inherent protection from viruses and worms. (That is, if I was after 12 beers or after a lobotomy. GNU/Linux and Mozilla have a metric buttload of security advisories too. And let's not forget that the words "rootkit", "to get rooted" or even "worm" come from the Unix world.)
But from spyware? Dream on. That stuff usually comes nicely packaged with an installer. And often hidden inside some other program's install, barely mentioned near the end of a 10,000 word EULA. (Yay! If you pay for our password remembering gizmo, you get our award winning spyware for FREE!)
The same Joe Average who cheerfully installs Gator on Windows, would just as cheerfully install it under Linux or MacOS. Even if he has to log in as root for that. (You did train him to only login as root to install stuff, so he'll do that.) Ditto for Mozilla spyware bars and plugins, if anyone bothered writing those. It's not like it doesn't support them or anything.
So basically you're bragging... what? That your great security advantage comes from being a minority noone gives a damn about? Well, gee.
Besides, Mozilla is buggy enough as it is. I guess even the scum at Claria aren't _that_ evil as to inflict even more pain on its poor users.
I'll go further and say there's another white elephant that everyone seems to forget: we've beaten the user into submission with idiotic confirmation dialogs for _everything_.
Do you want to save this password? Do you really want to cut this block of text? Do you really want to save your game? Do you really want to overwrite the file you were editing in the first place? (A ton of shareware/freeware programs love to ask nonsense like that.) Do you really want to view a non-encrypted page next? (Mozilla is actually idiotic enough to ask exactly that... even when I have explicitly typed the URL in myself.)
The user has been trained like Pavlov's dog, to press on "Yes" every time he/she wants to do _anything_. The problem isn't with "idiot users." The problem lies squarely with us idiots who've sold him those problems.
This is not only a Windows or a GUI problem. Think of the Unix command "rm". When was the last time you used it without "-rf"? Right. That's my whole point. Excessive confirmation just gets never used, whether by command line options, or by instinctively mindlessly clicking "Yes."
A better interface would involve as little confirmations as physically possible. Can we be safe without the barrage of confirmations? I believe we can.
E.g.,:
- don't ask the user whether he wants to delete a file. Make it possible to safely undo any file system operation. No, not just "last 100 MB deleted". If I did the equivalent of a "rm -rf/", a smart OS should allow me to undelete it all.
And not only for deletes. It should be possible to also undo those mis-clicks, which accidentally dragged and dropped a whole folder into another.
- don't ask the user whether to run a program in the browser. Make it safe to run it anyway. Java and its sandbox model are a great example of how to do it right. Unless explicitly granted extra permissions, a program can not access the local hard drive, listen to ports, connect to any server than the one it was loaded from, etc.
- if something is potentially dangerous enough to need a confirmation dialog, IMHO it shouldn't happen automatically in the first place. E.g., if executable code (e.g., plugins, drivers, codecs) need to be installed, let the user explicitly download and install them separately.
- more importantly: don't fscking nag. If the user explicitly refused something once (e.g., installing a certain plugin) or explicitly disabled something in the menus, remember that and _obey_. Don't hit the user with a barrage of popups asking again and again that maybe he still wants to reconsider.
E.g., if you disable ActiveX in IE, it will nag you with popups to re-enable it. E.g., if no, thank you very much, I don't want to install a certain plugin, IE will nag me to install it. (And worse yet, each time it will first hog the bandwidth downloading it and _then_ ask.)
I once had to install Windows 2000 on a box, and as Loki would have it, I had no Zone Alarm or Sygate Personal Firewall on a CD at hand. Just as Joe Average would.
So I could go download it somewhere else, or I could do a scapegoat installation just to download a firewall. I chose to just sacrifice an install to the gods of Hacking. I _knew_ I'd get hacked, but that was OK, since I'd reformat immediately after anyway. (Takes less time than whining on/. about MS security, btw.) Joe Average wouldn't know, and wouldn't reformat.
(And I'm not disappointed. It takes less than a minute to get my uplink bandwidth saturated with mysterious outbound packets.)
Still, it will serve to illustrate what happens after you get your machine 0wn3d by some l337 skr1p7 kiddi3.
So I decide to play with it a bit longer, and see what happens with a firewall and an 0wn3d machine.
I start the newly downloaded and installed Sygate Personal Firewall, and immediately it pops up a window telling me the name of the application _and_ what's it trying to do. I block it, and that's that. No more outbound packets. I can tell struggles long and hard to send crap, but it can't. Both its inbound and outbound pipes have been sealed shut.
I can now toy with that machine as long as I wish, trying to disinfect it. Again, which is what Joe Average would want. If it's _not_ a sacrificial install, but some machine where his resume and a few gigs of other important data is, Joe will not want it reformatted.
I can even surf the net looking for information on the trojan, safe in the knowledge that it's blocked. No need to pull out the network cable.
Whereas you tell me that Apple would have allowed it to open its own ports, as it damn pleases. Inbound or outbound, whatever. And not even told me about it.
Well, gee. Sorry, that's not the kind of security I'm looking for. Dumbing down a firewall to the point where it doesn't actually block anything, in the name of "user-friendliness" is _not_ the way to go.
"1. All-in-one = single point of failure when something breaks, and needs to be replaced."
Bullshit. See the airplane principle: an airplane with two engines has twice the probability of an engine failure. _If_ I was to assume that electronics have a high failure rate, I'd be more worried about having 10 modules than about having one.
But the failure rate isn't that high in the first place. My PS2 still works flawlessly. For that matter, so does my old Playstation. So does the Dreamcast I used to carry around everywhere. (Man, I loved that thing.) Admittedly, they've been well treated, but then I'd assume most people who are serious about watching DVD on one don't go around pouring Coke in them either.
And if it was to break tomorrow, wth, I got years of gaming _and_ DVD playing out of it. Methinks that's great value for the money.
And if we're talking reliability, let's talk a very related point: the advantage in using standard, proven components. The only ones experiencing any significant failures were the ones who went their own non-standard, non-converging way. Sega even got bitten hard by that, as their first units had massive problems with the still experimental "GD-ROM" drive. The ones who just went with a standard format, like DVD, even for convergence reasons, actually ended up with a more robust product.
"2. Generally speaking, all-in-one devices incorporate propietary technologies to promote lock-in and/or reduce 3rd party tech licensing costs for the company (SONY!!!!)."
Bullshit. You don't need proprietary DVDs to play on a PS2. Any old DVD will work.
Au contraire, if you want "proprietary" and "lock in", look no further than Nintendo. Or Sega. Where Sony put a bog-standard DVD-ROM drive in the PS2, both Nintendo and Sega went and made their own obscure proprietary CD formats.
"3. Quality of stand alone components is usually much higher. Think stereo equipment."
Bullshit. Or at least missing the point. Not everyone is a rich consultant, who must have some $3000+ designer module even to play CDs.
Sony's components in the PS2 are perfectly on par with most consumer grade stuff, and actually better than the cheap on-board sound you'll find in most PCs. And you get all that for a mere $15 extra. For most people it's superb bang per buck.
"4. Modularity = more cost effective upgrade path."
Bullshit. More cost effective than $15 for the DVD playing capability? You must be smoking some really good shit.
"5. All-in-one = usually more complex than individual devices. Stand alone means you can learn and understand the functions fully before moving onto the next component. Sometimes the 'role' of a device is confused when it is consolidated. e.g. Does 'play' mean play the.mp3, the CD, the DVD, or the video game???
Bullshit. Have you even seen Sony's remote control, or are you talking out of your ass?
Hint: It's a standard remote control, just like those you'd get with a standalone DVD or VHS player. Anyone who can't figure that out, is probably too stupid to use any other remote control too.
Plus have you even seen a console? It only has one drive. How the heck would anyone get confused about "Does 'play' mean play the.mp3, the CD, the DVD, or the video game???" It plays whatever you've just put in that one drive. Just like any other console.
And it has auto-play, just like Windows. So you can just see it on the screen if a game started, or you get a DVD's menu.
Briefly: anyone stupid enough to have trouble with that, would probably be too stupid to figure out whether they're playing Donkey Kong or Castlevania on their Nintendo 64. Or for that matter whether they're playing Doom 3 or Solitaire on their PC. I.e., at that point the problem isn't the complexity, it's just a user with single digit IQ.
"6. All-in-one convergence not always a logical combination. Digital camera cell phones? mp3 playe
Guess one can never get enough of "waah! but what if someone had patented sex!" kind of fallacies on/. Was starting to get withdrawal syndromes after through a whole weekend without reading one;)
But OK, let's play that game. Let's talk about paintings:
1. They'd more likely have to patent a device or method to make those paintings. So someone might have got a patent on something new like flinging colours at the canvas, but then someone else might just as well get the same result (or close enough) by using the old methods (using a brush). For which plenty of prior art existed.
(Just as this patent doesn't prevent you from having a lit case by other means than what Apple patented. You can still have your old cathodes or LED fans.)
2. For that matter, it might have stimulated someone to try more new stuff. So we might have 3-4 times more styles in the same period. Which is the whole purpose of patents: to stimulate researching _and_ publishing your research.
(And you could say the same about the situation at hand. We've had _years_ of noone even trying something more original than yet another LED fan or cathode behind acrylic window. By now every kiddie has one of those. So if it takes patents to get out of that loop and have a more original case, seems to me like a benefit of patents.)
3. Patents are not for ever. Copyrights amd trademarks do get extended. Patents expire no matter what.
I.e., if you talk about a 100 years interval, you may notice how the 20 years covered by a patent is only a fifth of it. I.e., combined with the previous point, we'd probably have a helluva lot more art choices after 100 years.
4. Patents encourage publishing your results, as opposed to keeping everything super-secret. Art is a bad example there. But there are a ton of technological processes that one could have kept secret. Or which _are_ being kept secret. Patents encourage companies to share this information with the rest of the world instead.
5. Patents get licensed all the time. I'm sure that if someone absolutely needed to do something impressionistic before the patent expired, they could have negotiated a license.
(And in this case, if IBM or MS absolutely needed to do their own lit cases, I'm sure they can persuade Apple.)
I don't know even about very young children. I'm getting the idea that the only ones who make "educational games" are the ones too fucking unskilled to make a proper video game.
Can games teach people stuff? Well, yes. I've learned a lot of history stuff from games like Europa Universalis, or to a lesser extent Civilization. Or at least got the curiosity to read more about that from other places.
Or "Die Gilde" ("Europa 1400 - The Guild" for you 'merkins) gives you a historical report of what happened IRL in that year after each game year. I've learned more late medieval trivia from it than from any other game.
But here's the scoop:
1. It must be fun as a _game_. Civilization was a bestseller in its own right. It didn't need to masquerade as "educational software" to get any sales at all. Ditto for Die Gilde, at least in Germany. Europa Universalis has a steep learning curve, but also got quite a few people addicted on its own merits.
2. Don't lecture or preach. It must first and foremost be a game, not a piss-poorly disguised beating people up with a clue stick. People instinctively resist being lectured.
3. Don't be patronising. Stuff that basically says, "see, we know you're a fscking retard who doesn't know how to put stuff in a fridge. We also know you're an idiot who can't figure out the cheapest crap to buy." serves no purpose other than humiliating someone. If anything, it'll make them resist the lecture even more.
And I'm thinking the same could be applied to software for small children. A game should be written to be first and foremost a _game_, and only incidentally also education.
E.g., there are a ton of _fun_ ways to make someone exercise their maths or logic skills. Economic sim games have done that for ages. Puzzles are also a good means to that end. (And god knows even the worst maths puzzle is still better than yet another "jump puzzle".)
So it's not like they _have_ to be crappy _and_ patronising games to be educational. It's just that the people making them seem to be into patronizing their gamers. And in most cases also utterly unable to make a proper game anyway.
Well, the thing is, the ads in magazines are very targetted. They fit the narrow scope of the magazine, and if I'm blowing some money to read that mag, chances are I'm interested in that topic.
E.g., if I'm reading a hardware magazine, chances are the ads will be hand-picked to be about hardware. If I'm reading a Java magazine, it'll probably have _something_ to do with Java. (Even if sometimes as little as "our new Ultra-Expensive Series (TM) servers run Java very very fast.") Etc.
It works on the Web too, not only in mags. E.g., I bought City of Heroes because several online comic strips ran, well, strips about it. They weren't intended as ads, but they worked as such admirably. It was as targetted as it gets: I was reading _comics_ about _games_, so it's a safe bet that both rate high on my interests list. So when they mentioned a game about comics matched both.
However, the thing is: they're hand-picked.
I don't think simple keyword matching can come even close to that. For web advertising to truly work, it would have to be a lot more targetted and relevant than it currently is. It needs a human to do the matching between ads and content (or at least the site's general direction.)
Well _if_ targetted marketting was actually smart, and not just word matching, that and probably a lifetime supply of game ads. Since mostly I'm on game related boards anyway.
Of course _really_ smart targetted marketting would offer me an auto-trolling program, so I don't have to type all that stuff. One Click Trolling. Someone ought to patent that.
Of course, being just keyword matching, I generally still get only irrelevant stuff.
For example I check out the lyrics to a gothic song about death and funerals. (Standard goth stuff, really;) And, presumably because it mentions "eternal sleep", I get an ad for sleeping pills. Erm. I really don't think they meant that ad to be targetted _that_ way. Just because I listen to gothic stuff, doesn't mean I want to commit suicide, dammit.
If you actually started recording surfing habits through these keyword matchers, it would get even more funny. For example, I'm sure that based on the words on those gaming boards, someone could conclude that my interests include:
- archery (what with all those bowmen/crossbowmen/etc in strategy games)
- horses (stable buildings in games, the occasional historical debate about stirrups and knights, etc)
- nazis (well, they're in every other game)
- voodoo (well, zombies anyway. They're a popular target in games) and generally magic (RPGs. Nuff said.)
- stealing (can't have an RPG without thieves)
Etc.
I shudder to think what kind of ads I'd get if someone actually collected that data.
What did you want me to do? Write a whole tome worth of everything that ever was wrong with Java client side? I did say "That's just a small slice of the many ways in which Sun started it on the wrong foot."
Believe it or not, I've been in client side application projects. I also have Eclipse open right now.
In fact, Eclipse is the prime example of how Sun started on the wrong foot. SWT (the widget set used by Eclipse) is IBM's work, while AWT and Swing are what Sun had in mind. The words "dumb and dumber" come to mind, when I think about AWT and Swing.
Want one thing that an actual client complained about? We had to program the same very small app in C _and_ Java, on Windows, Linux _and_ MacOS. Well, the whole package were more apps, but this was the only part which had to run on all desktops.
The client's first comment? "Whoa! Why does the Java version need 16 MB of RAM, when the C one is less than 1 MB?"
Another project was a bigger Swing "thick client" database program. In Swing. You know what the first complaint was? The look and feel. Inconsistent fonts, inconsistent keyboard shortcuts, etc. Sun's "Windows" Look and Feel is a sick joke. We ended up doing more work to write our own look and feel, just to get the program to look like a native app.
Yet another way in which SWT got it right, while Sun screwed up.
The second complaint? Performance and memory use. 'Nuff said.
Also, believe it or not, bigger companies have tried moving to Java on the desktop. It was every company's wet dream to suddenly only have to maintain one version, and work on every single OS. Some of which they didn't even support before. Everyone tried it.
E.g., Corel at one point wanted to move their whole office packet to Java.
Why do you think Microsoft got so scared and wanted to kill Java? Precisely because of this. It looked like all of a sudden everything will be multi-platform, and the "but this and that runs only under Windows" advantage will evaporate over night.
What happened to all those projects? They flopped. They ended up slow, bloated and more unstable than their Windows counterparts. So everyone moved back to C/C++.
If that doesn't count as a monumental flop, I don't know what does.
Doing at least basic image processing on photos taken with their digital camera? You'd be surprised how many moms and pops at least try their hand at photoshop nowadays.
Some form of media encoding? You don't have to be a genius to run a file through DivX, you know. Or to rip your collection of music CDs to MP3s, so you can put them on an iPod.
Antivirus scan kicking in? You haven't really learned what "slow" is, until you've witnessed Norton's or McAffee's scan kicking in. In fact, much of the popularity of, say, Intel's HyperThreading isn't because people have heavily-multithreaded apps, but because you can still reasonably use that computer while virus scans and such kick in.
(Not that their real time scanning is any better. Watch the same directory full of small files take under a minute to copy to a file server, without Norton Antivirus. Watch it take more than half a fscking hour with Norton Antivirus on.)
And so on and so forth.
And even games aren't that unheard of among the less computer-savvy people. Just because someone isn't a die-hard smack-talking rocket-jumping Doom 3 player, doesn't mean that everything they'll ever play is Solitaire.
Java is pretty popular on the server side, but client side it was always one monumental flop.
As applets go, for example, nowadays the whole program-inna-browser market is owned by Flash, followed by ActiveX. And for good reasons.
Starting with the fact that Java 1.0 was indeed a slow piece of crap for anything but the most trivial applets. Try displaying a complex table without a JIT, and you were talking about response times you could measure with a stopwatch, not with System.currentTimeMillis().
The initial lack of support for packing everything in a jar didn't help that cause either. Downloading 50 classes as separate files isn't particularly fast. And that's a very small project.
And for all the multi-platform hype, wasn't particularly portable either. If you tried running even a trivial AWT applet on different platforms, you wouldn't even get the same events. Or for something which required you to give a size in pixels on the web site, you wouldn't even get the same font sizes.
And by the time it caught up... meh. Flash is _still_ the better choice.
Not the least because of download size. Sun now includes all the crap they could think of as standard libraries. Do I need an XML parser to make a simple game applet? Not really, but Sun wants my users to download that crap anyway.
(No, it's not a made up problem. I've had modem users tell me literally "whoa, I'm on dialup. Is there some smaller version I can download?")
That's just a small slice of the many ways in which Sun started it on the wrong foot.
We have a thread which basically only says "let's try to attract them to CS. Who knows, they might even like it?"
We're _not_ talking about enforcing 50% quotas. We're _not_ talking about punishing people for not hiring women. We're _not_ talking about any extra incentives to hire women either.
It's _only_ about introducing some people to programming. Just a stupid summer camp with computers. That's all.
Yet someone feels a need to throw a tantrum about "feminazis" and "man-haters", and inherent gender differences, and how it's ok to keep them in underpaid underling roles. 'Cause, you see, those are important too.
Someone for whom the very idea of women even being invited to a stupid CS-themed summer camp, is already enough to trip a whole rant.
You're talking about conducting an interview to see who's got the skills. (Much as I believe objective CS interviews to be on par with Santa Claus and the Tooth Fairy, but at least the intention is there. Or the wishful thinking.) Whereas he is going on a tantrum at the mere idea of women being introduced to computers.
Sorry, that _is_ discrimination.
Dunno about you, but I'll call him a flaming idiot all right.
Dunno about you, but I hope such bigotted idiots aren't allowed anywhere _near_ a hiring position. Someone who's mortally offended by the idea of even letting women near a computer on a summer camp... there's no way in heck I can believe that such people can be objective and biased in a job interview. More likely he'll just go offended by the idea of women in non-underling roles again.
Look, I'm not demanding quotas or anything. (Although, come to think of it, it couldn't make hiring in this industry more of a joke than it already is, anyway.) I'm not saying to prefer hiring women or anything.
I'm just saying: start with an open mind, and at least try to keep prejudices out of the way. And FFS, don't throw tantrums to the mere idea of a CS summer camp for girls. That's all.
And if you're already doing that, well, you're not the kind I was calling an idiot anyway.
Well, where are the good male programmers? About 3 out of 4 can't program at all. They just don't have the intellectual capacity for it. More than 2 out of 3 don't even know the language they've put on their resume.
I work with some people which don't even understand what "call by value" means, or what a hash table is. At all. And some competent people, but more of a coincidence than any merit of whoever did the hiring.
So I'll go further and say: where are all those interviews which actually determine someone's skill? I've yet to see any job interview which was even a test of aptitude. At all. It invariably was a "prove to me that you're willing to bullshit your employer with a straight face" contest.
Being a honest guy, and having enough real stuff on my CV anyway, I say just that. On the other hand, invariably I can tell that I could just as well have said that I was the IT Grand Vizier to the Sultan of Cyberia, and would have got the job just as well. And invariably in the same week someone completely incompetent got hired as well.
Often complete with bullshit questions like "what is your biggest deffect?" No, it never ends up testing someone's honesty. It's just an exercise in creative bullshitting.
Look even at some examples cited on/. that passed that test. Invariably stuff like "oh, my biggest problem is that I'm totally focused and excel only at what the job description says". Only phrased in a way that snowed a clueless HR droid. Well, gee, anyone whose biggest problem is being a great professional, must be more perfect than Jesus.
Or we've had article right on the front page of/. pointing to people who are proud of being arbitrary just because they can. E.g., dropping CVs just because they don't like the email address. Or various other examples of an insecure incompetent having power trips along the lines of "fsck you buddy. I don't need you. You're the peon grovelling for mercy at my feet. So let's see you convince me why should I waste my royal time and bestow the undeserved favour of a job upon you."
I've yet to be asked some relevant question. Like, say, "when would you _not_ use <currently hyped and fashionable technology>". (Notice the "not" there. Any monkey can learn the hype, but it takes some skill to do actual design.)
So where are all the interviews which objectively prove who's the better professional? No, really. I want to know. Maybe I'll apply for a job there next time.
So for me to believe that a job interview objectively proved some gender/race/nationality/whatever differences, I'd first have to believe that it objectively proved _anything_. As long as it's mostly just a "which of these guys I like the most" contest, it'll at best just be an exercise in reinforcing a pre-existing prejudice.
Disclosure: I'm male. I also believe that respect ought to be awarded for _personal_ merits, not by just being born into the "right" group.
And basically what you're telling me is along the lines of "underpaid underlings are signifficant too, so it's perfectly ok to force women and minorities into those roles. And therefore it's ok to pay them less."
Gee, what a flaming idiocy.
You know what? It's not the PC crowd that are the illogical zealots, it's the relics from the past like you. Idiots who think they should be paid more just because they happened to be born white or male or whatever.
Well, let me break some grim news to you: if your _only_ merit is that Y chromosome, you're a fucking waste of air. That chromosome alone doesn't even make you necessarily smart enough to flip burgers at McDonald. Much less make you qualified for science or engineering. Much less for CS.
If you want any respect or a better paid job, show me what _you_ personally have done to deserve it. Not what random thing you've got at the chromosome roulette.
And if you actually need to blame your problems on women, blacks, jews, corporations, international conspiracies or whatever, you've just earned a hefty dose of disrespect from me. Means you're not only an incompetent piece of waste, but unable to take responsibility for your own failures either.
Let me spell it out: anyone who's anywhere _near_ competent in their job or life, doesn't need a scapegoat to put the blame on. Sure, bring in the women, blacks, indians, or the mongolian hordes for that matter. I know I won't lose my job in the process. Or will find an equally good one quite easily.
Do _you_ feel threatened by "feminazis"? Excellent.
When I was in the teens, I also had the idea to make a virus. Mind you, back then it meant file or boot sector virus, not an exploit over TCP/IP. We're talking the 80s.
Coding one was actually a very trivial exercise, and I like to think I did a fairly good assembly job of keeping it under 512 bytes, so it wouldn't be too obvious. It also worked.
But here's the catch: even at that kind of age, I was perfectly able to know it was wrong. That virus was only tested and demonstrated to a couple of friends in a controlled non-networked environment, then destroyed. It also carried no malicious payload anyway.
Teenager or not, I think everyone I knew was perfectly able to distinguish the moral difference betweem a "let's see how this works" exercise, and releasing a plague. Some of them might have thought it would be way cool to cause that harm, but I don't think anyone was incapable of discerning that it was wrong.
And let's forget viruses. We trust teenagers with a lot of other stuff.
E.g., depending on the country, you may be allowed to drive at 16 years old. You are trusted to drive a dangerous ton of metal, and not suddenly go running people over.
E.g., in a lot of countries you can be conscripted at 18 years old. And then given an assault rifle and live ammo, and left alone with it for hours, guarding some corner of the compound. Someone incapable of discerning consequences would start shooting at people to pass the time away. But in practice they don't.
So, well, gimme a break. If teenagers can be trusted with an assault rifle or a car, I do believe that an 18 year old is also very capable of understanding what a virus is.
I'm such a geek. Normally I don't really give a damn about what people think of what I wear.
(It's decent. I'm not wearing a crotchless S&M outfit to work, or anything. If anyone is fundamentally offended the sight of a clean pair of jeans, they're just stupid. And I have better things to do than worry about stupid people.)
However, in this case we're talking a court of law. You don't want to piss off the judge who might, on a whim, give you a suspended sentence or community service or send you behind bars for a few years.
You _don't_ want to look like an unrepentant "fuck you all" rebel to the judge. You don't want to look like you're damn proud of what you've done. (Which is the impression that such a "Big Daddy" t-shirt would have given even me.)
I'm not even saying he should have worn a suit and tie or anything. But, you know, even if you're gonna wear a t-shirt, make it a plain one.
I mean, geeze, wear that t-shirt to school. Wear it at a party. Wear it even to a job interview if you honestly don't give a damn about the outcome. But a court of law is more serious: unlike a job interview, you can't just try again somewhere else.
Basically all I'm saying is that there's difference between not caring about stupid people, and _being_ the stupid one. Freakin' big difference.
Actually, while QWERTY wasn't designed to slow people down, it _is_ designed to avoid jams.
The thing is, the contraption consisted of (more or less) a semi-circle of thin levers, each with a little hammer with an embossed letter on it. All were aimed at the same position on the paper. You press a key, and purely mechanically the lever would swing the hammer at the paper. (Well, actually, at the ribbon.)
Also, because it was a purely mechanical contraption, the cheapest and most reliable way to build one was: keys that are close on the keyboard, would also activate levers which were close to each other.
Jams would happen when two close enough levers would be activated at the same time. Or close enough. The closer the levers were, the more likely you'd get a jam. (Again, purely coincidentally, this also meant "the closer two keys were".)
E.g., pressing "Q" and "P" at (almost) the same time would never jam. They swung from opposite directions, and it was pretty much guaranteed that one hammer would simply hit on top of the other. E.g., "A" and "S" at the same time (e.g., while typing "ASSASSIN") would pretty much always jam.
So basically, QWERTY:
1. was just supposed to prevent jams. (Which cost more in typing speed than a couple ms worth of more finger movement.)
2. was not designed to do anything to typing speed as such. Neither maximize it, nor minimize it. Whatever typing speed difference it produced, it was "side effect", rather than "goal". (And, again, a lot of it came from jam prevention rather than anything else.)
3. the _only_ typing speed consideration it received at all, was a rigged tech demo. Ever wondered why the "QWERTYUIOP" row? Because the rigged tech demo was basically "Look! I can type 'TYPEWRITER' quickly! It must be an optimal layout!" Hence all the letters in the word TYPEWRITER had to be on a single row.
(Hardly a scientific study, but PHBs bought it anyway.)
Furthermore, I'd point out that:
A. It did a piss-poor job even at spacing common letter combinations apart. E.g., even in their tech-demo "TYPEWRITER" they have letters which are near each other: "TY", "EW", "ER", and thus prone to jamming. "W" and "R" aren't that far apart to be jam-proof either.
B. if you've ever used one of those purely mechanical typewriters (no, some electronic thing doesn't count), you'll notice that typing was a different exercise on those. It involved keeping your hands above the keyboard and hitting the keys pretty hard. At the very least it's _not_ the same RSI prone position you'd use on a normal PC keyboard.
C. a PC keyboard doesn't jam.
D. Even if you do type the wrong letters on the PC, the cost of errors is next to nil. Correcting a mistake was a _very_ time consuming operation on a mechanical typewriter, since it involved physically erasing or covering printed stuff with white paint. By comparison, hitting backspace on the keyboard costs a small fraction of a second.
Etc.
So basically I'm saying that the considerations from which QWERTY was born, not only were imperfect to start with, they bear exactly _zero_ relevance to a computer keyboard. That QWERTY still works well, is more of a testimony to the fact that people can learn _any_ keyboard layout well enough, than some inherent advantage.
QWERTY, Dvorak, even alphabetical order, IMHO you probably just type faster on whatever you have more exercise. That's all.
The thing is, far from Apple having a revolutionary new model, they're just stuck at a model that existed ever since the first computers.
In fact, in the beginning, there was no software market at all. When you got a computer, say, from IBM, the _only_ software you could ever hope to get was also from IBM. That included the OS, assembler, compilers, utilities, _everything_.
And chances are you couldn't plug in any peripherals that your vendors didn't sell, either. Some vendors went as far as to patent the connectors, to keep other people from undercutting the price of their ridiculously high priced peripherals.
That kind of a lock-in isn't new, and isn't to keep the customer happy: it's most marketroids' wet dream come true. It's pretty much like having a captive market, sorta like the 17'th century colonies. Or like having serfs paying the yearly Sun/IBM/DEC/whatever tax.
Except this time the fear was not of the King's army, but of the costs of rewriting everything to a new platform.
Even when Unix came and (theoretically) gave the proprietary OSs a kick in the pants, ever wondered why the Unix fragmentation happened? (Which in turn paved the way for Microsoft's defeating them all.) No, seriously. Because noone actually wanted to have a compatible platform, which would break their customers free from the vendor lock-in.
A truly portable Unix and tools, would have allowed you to just take your programs off that DEC mini and move them to a Sun. Or viceversa. Pretty much forcing a competition strictly on price/performance, instead of a cost artifficially skewed by the migration costs.
So what did everyone do? Made their Unix version slightly different. Made sure that although theoretically they were all Unix and all ANSI C, you couldn't really just take your programs and scripts and move to a rival's computer.
It's not even limited to whole computers: see how 3DFX guarded Glide jealously until right before the end. They wanted people locked into needing a 3DFX card, instead of free to get anything.
Or not even limited to computers at all: see, for example, how Apple doesn't want Real being able to sell tracks that can be played on the iPod. They want you locked into having only one vendor of DRM-ed music for your iPod.
So basically, again, Apple isn't really doing anything new. It's just doing the same old crap of trying to lock the customer in.
And if that's the great liberation from Microsoft's tyranny... well, count me out. I'll stay here where at least I still have _some_ choice. I can at least understand the Linux crowd to that end, since Linux _is_ free of proprietary crap and vendor lock-ins. But Apple? Gimme a break. They're twice as bad nowadays as MS and IBM put together ever were.
Anything worth protecting in a company, they'll patent. You know why? Because patents give you a 20 year unconditional monopoly, while "trade secrets" in a product you can just open up and copy are a toss anyway.
Trade secrets are valuable when you can actually keep them a secret. E.g., some secret stock picking algorithm that runs only on your computer, which isn't even talking directly to the web server. They just share a database server, past 5 firewalls and 2 armoured doors. Someone would have to actually break into the building (or bribe the sysadmin) to get to that.
But in a product which you ship and anyone can take apart and/or reverse engineer? Gimme a break. There's a reason why even software corporations want patents instead: because they're a far safer bet than pretending it's top secret.
The way Seagate's GMR heads work, is not only the same as everyone else's work, but you could just take the drive apart and find out for yourself anyway. The way their caching and TCQ algorithms work, you could just disassemble the firmware and see for yourself.
So IMHO Seagate is just having a knee-jerk reaction, whether they have anything to protect or not. They're just doing it because everyone else is doing it, not because someone sat down and analyzed what secrets they have to protect.
Same as we've seen during the dot-com scam. People made you sign NDAs to protect some super-secret plan like "umm... we're making a web site. With a forum. And... umm... yeah, I know, lots of Flash, bright colours and 1 meg of.gif roll-overs per page."
Well, geee... surely noone else would have thought of that on their own. Noone except the other tens of thousands of equally clueless dot-coms, that is. And surely noone can get that idea by just seeing the bloody web page. They surely _have_ to steal your employees to figure that out.
Yet they made you sign tome sized NDAs. One I had in front of me wanted to prohibit me from ever competing not only with them, but also with their clients, or their clients' clients' clients. In any way. Taken literally, if they ever made some ad for a news site, I'd have been prohibited to ever blog, 'cause that would sorta be competing. If one of their clients was a detergents corporation, I'd have been quite literally prohibited from selling soap.
And, again, to protect what great trade secret? That they want to make a web site. With flash. And ads. Gee, that's so revolutionary, non-obvious and worth protecting.
So personally I'd very much like to see this idiotic cult of the NDA die already. My message to these employers would be: if you have something worth defending, get a patent already. If it wasn't worth patenting, go fsck yourself and get off my case already.
I don't think that structure or strong typing precludes good design or tests. You can have it all.
Tests are good and fine, and everyone ought to use them. Indeed.
But keeping me from passing the wrong arguments to the wrong function is another source of bugs eliminated. A source of bugs which can sometimes be rather insidious, since parameters can come from anywhere and get passed through 20 function calls. They can come from a user who (sometimes even deliberately) enters wrong data, the web designer who did this small change in the template and replaced the numeric ids with string ones, or the poor maintenance programmer who has a deadline to fix something he doesn't even understand.
And let me stress the last part: maintenance is the bigger problem, not writing the code. It's also the under-manned part: while when coding you could get only 10,000 lines per person, in maintenance one single person can well get a 100,000 or even 1,000,000 line project alone.
That person sees the program like through a keyhole. It's like seeing the Sixtine Chapel fresco for the first time... through a pinhole. You can scroll the view, but you see a couple of square inches at any given time. Good luck seeing the big picture. Good luck remembering it.
But when you're doing maintenance you're not always even allowed the luxury of getting the big picture. You're just expected to fix this one small problem, and fix it yesterday. Just quickly find the line that does that, and change it already.
Which brings problems like:
- where the heck _is_ the code that does that?
- where does it get its parameters from?
- are those parameters what I think they are? (E.g., does it always get a ClientDO object? Can it be null? Can it get some other data type instead?)
- I see FunctionX and FunctionY being called. Which in turn call functions A, B, C, D _and_ E. What _are_ their parameters? Am I sending the right data?
(E.g., I see it expects a Collection object. Will _any_ collection do? How do I know it doesn't get cast to ArrayList down the line. Don't laugh, I've seen code like that.)
- is my change breaking anything else? (E.g., passing the wrong data type to a function.)
And at that point _any_ help is most welcome. Good structure and strong typing and unit tests can give that maintenance programmer vital hints. Hints which otherwise he/she would have to spend weeks digging in code for.
That's really when that stuff is the most valuable. Sure, when coding you can often do without them. It's maintaining someone else's code that's the real kick in the nuts.
Re:philosophy and science have always been linked
on
The Unknown Newton
·
· Score: 2, Insightful
Stuff like Heisenberg or Einstein's relativity are, no offense intended, most quoted as "science doesn't have all the answers" by people who don't understand them at all.
Heisenberg itself doesn't just say "buggrit, you just can't know where the electron is", it gives you a formula there. You can actually calculate stuff and build practical stuff with that knowledge.
(E.g., the Zenner diode. E.g., Heisenberg may well be the reason why CMOS miniaturization comes eventually to a halt. And it offers you the formulas too to know what voltage you need to keep enough electrons there, for a given transistor size and shape. Etc.)
Second, it too can be observed experimentally. It's not used as way to throw up your hands and say "the electron does whatever God wants", but to understand and _predict_ where the electron can go. And within what margin of error.
Is there a ton of stuff we don't know about yet? Most likely. Are there limits to our theories than we don't know about yet? Almost undoubtedly.
But we'll cross that bridge when we come to it. And then we'll have an even more complete theory, and more advanced experiments to test its limits.
And in the process we'll be a lot more advanced and able to build more useful stuff, than if we had all just thrown our hands up and said "buggrit, it all happens because God wants it to." Well, sure, that may be so, but God's will alone hasn't produced any predictable results so far: you can't build a transistor based on it
Slashdot Mirror
Given the tons of garbage on the net, and how half the search topics give you 10,000+ blogs, clan sites, message boards, etc, if you can actually discern between the right info and the bullshit, then IMHO you already knew enough to write the paper on your own.
As a random example, if I were to write a paper about the crusades, based on info on the net, I'd be faced with a deluge of conflicting information. I'd have to discern between posts and pages by people who knew their stuff, and bogus pages and posts by gamers who got their facts from games like Lionheart.
And I doubt that any professor would appreciate if I handed in a paper saying that Richard Coeur de Lion (Lionheart) opened a rift to hell and unleashed demons upon earth. And then allied with Sultan Saladin to fight them off.
Hey, it's in the game Lionheart, so it must be true.
Or: were the crusades started really by Peter the Hermit, to free the holy land, or was it all the Pope's plan to help the Byzantine Empire against the Turks? Would the crusades have still started if the Byzantines weren't starting to take a beating? You can find both points of view presented as historical fact on the net.
Or: which crusade conquered and sacked Constantinople? Was it the fourth, the fifth, or which? And why did that happen? Again, you can find the most conflicting information on the net.
A lot of it is simply conflicting historical data, because different chronicles back then had different biases. But about 75% is simply people taking out of their ass. Just because, you know, it's the Net and you can publish _anything_.
It gets even sadder when it's about programming. Every burger flipper can get hired as one, and God knows every other of them has got to re-publish on the net some "cool tricks" he's found on the net. Half of which don't even work, or are "optimizations" which actually make the program run 10 times _slower_.
So basically it seems to me safer to just read the book and do your own research. By the time you're actually competent to discern the gold from the bullshit on the net, you've already done your research and are more than competent to write your own paper.
There's another reason why your "adding a VCR to an existing TV setup" example is the prime example of what's wrong with computers today.
If you added the VCR wrong, you get feedback. It's obviously not working. You know you need to try again, or get help from someone who knows.
Whereas an insecure WiFi setup gives you no hint at all.
In the quest to _seem_ easy to use, but without actually having to invest in real ease of use, we're just covering up the problems and hoping that noone notices.
Everyone wants their device or program to look like it's just a trivial plug-and-play affair, so what do they do? Maybe actually invest in making it so? Nah, we'll stick to cutting corners instead. We'll make it plug-and-run-wrong (e.g., insecure by default or with a hard-coded admin account) and hope the user doesn't notice it's broken.
And when he does notice that for half a year he's been running a porn server _and_ a spam server _and_ his computer and connection were clogged... we'll just call him an idiot.
In reality, that's simply broken design, not the user's fault.
If TV manufacturers worked that way, instead of giving you digital tuners and a remote control, they'd just default to showing you one preset station so you hopefully don't notice when you've tuned it wrong. (And, hey, just think of the monopoly possibilities and raking the big ad bucks.)
"1. When was the last time someone 0wn3d your TV or VCR?"
Actually, I hope you do realize you've just proven the other guy's point. That computers are such a fragile tool, and for a lot of people they can cause more grief than good, is precisely the _problem_.
What Joe Average wants -- or for that matter what _I_ want -- is something that just does a certain job, with a minimum of fuss. Yes, like a TV or a VCR. If I want to read my email or play a game or whatever, I should just get straight to doing that, instead of having to babysit and secure a piss-poorly made tool.
"2. More complex systems require more complex instructions."
No, it's just a case of letting the idiots run the show. Plain and simple.
Other tools started complicated to use too. Owning a car used to require either being a skilled mechanic yourself, or being rich enough to pretty much hire one full time. Getting an early radio to work, or tune it to a station, was a time-consuming pain in the butt. Etc.
But you know what? Someone in those industries actually cared for the customer. (Or just about the bottom line. Competition is good at that.) Instead of whining about idiot users who can't even learn to use a radio right, they gave you channel presets, auto-scanning for stations, remote controls, and other such.
That's really the only problem with computers today. That instead of asking "how could we make this easier for Joe Average?", we're whining about how Joe is an idot and a luser who can't learn doing things our arcane way.
E.g., if we're talking about wifi, it would be a no-brainer to:
- have a nice wizard interface and walk him through securing the thing.
- make sure that security is enabled by default, and that Joe has to explicitly disable it, if he _really_ wants to run a public "download porn and warez anonymously" service.
- If the device has a default admin username and password, explicitly ask him to change it.
- But what if Joe forgets the password? No problem. Don't fscking have an unchangeable one hardcoded in firmware. Provide an easy way to change it, but which requires physical access to the device. E.g., have to open a lid and press a sunk reset button. After which again, make him change it.
Etc.
See, it didn't even require that much thinking.
But no, instead we'll just whine about how Joe is an idiot luser. Although it's not Joe who's the idiot there.
"3. Adding a wifi router to an existing computer setup is more akin to adding a VCR to an existing TV setup."
I'll direct you to your own point 1: when was the last time someone "0wn3d" your TV after that?
Or if we're talking unneeded complexity, when was the last time you had to become a security expert to add a VCR? Did you have to just know how to generate and share keys on them? And did you need to find that out on your own?
Well, my real curiosity is "why the heck does Canada need _cooling_ the air?"
Look at a world map and see where Hungary is, and where Canada is. Hint: a lot more to the north. It's up there with Siberia.
Or with Scandinavia, but the difference is that Scandinavia has the warm Gulf Stream. Canada doesn't. As I've said, the most apt comparison is Siberia. Another way to put it would be "fucking cold."
So why would anyone need to cool the air is beyond me. Might as well enjoy that warmth while it lasts, IMHO.
Seems to me like what they'd _really_ save energy with, is if they found a way to naturally warm the rooms in winter, rather than on cooling them in what little summer they get.
Then again, I'm sitting at 30 C in an office without air conditioning, and I think it's very nice. So your mileage may vary.
Problem solved only because noone gives a damn about something which has maybe 1/1000 of the market. Mozilla on all platforms is already a spit in the bucket. Mozilla on MacOS, well, that's got to mean... what? You and 10 other users? Hardly a market worth programming for.
The fact is: nothing in MacOS X, Mozilla or even Linux would protect you from spyware.
_Maybe_ I could be persuaded to believe it gives you some inherent protection from viruses and worms. (That is, if I was after 12 beers or after a lobotomy. GNU/Linux and Mozilla have a metric buttload of security advisories too. And let's not forget that the words "rootkit", "to get rooted" or even "worm" come from the Unix world.)
But from spyware? Dream on. That stuff usually comes nicely packaged with an installer. And often hidden inside some other program's install, barely mentioned near the end of a 10,000 word EULA. (Yay! If you pay for our password remembering gizmo, you get our award winning spyware for FREE!)
The same Joe Average who cheerfully installs Gator on Windows, would just as cheerfully install it under Linux or MacOS. Even if he has to log in as root for that. (You did train him to only login as root to install stuff, so he'll do that.) Ditto for Mozilla spyware bars and plugins, if anyone bothered writing those. It's not like it doesn't support them or anything.
So basically you're bragging... what? That your great security advantage comes from being a minority noone gives a damn about? Well, gee.
Besides, Mozilla is buggy enough as it is. I guess even the scum at Claria aren't _that_ evil as to inflict even more pain on its poor users.
Bingo.
/", a smart OS should allow me to undelete it all.
I'll go further and say there's another white elephant that everyone seems to forget: we've beaten the user into submission with idiotic confirmation dialogs for _everything_.
Do you want to save this password? Do you really want to cut this block of text? Do you really want to save your game? Do you really want to overwrite the file you were editing in the first place? (A ton of shareware/freeware programs love to ask nonsense like that.) Do you really want to view a non-encrypted page next? (Mozilla is actually idiotic enough to ask exactly that... even when I have explicitly typed the URL in myself.)
The user has been trained like Pavlov's dog, to press on "Yes" every time he/she wants to do _anything_. The problem isn't with "idiot users." The problem lies squarely with us idiots who've sold him those problems.
This is not only a Windows or a GUI problem. Think of the Unix command "rm". When was the last time you used it without "-rf"? Right. That's my whole point. Excessive confirmation just gets never used, whether by command line options, or by instinctively mindlessly clicking "Yes."
A better interface would involve as little confirmations as physically possible. Can we be safe without the barrage of confirmations? I believe we can.
E.g.,:
- don't ask the user whether he wants to delete a file. Make it possible to safely undo any file system operation. No, not just "last 100 MB deleted". If I did the equivalent of a "rm -rf
And not only for deletes. It should be possible to also undo those mis-clicks, which accidentally dragged and dropped a whole folder into another.
- don't ask the user whether to run a program in the browser. Make it safe to run it anyway. Java and its sandbox model are a great example of how to do it right. Unless explicitly granted extra permissions, a program can not access the local hard drive, listen to ports, connect to any server than the one it was loaded from, etc.
- if something is potentially dangerous enough to need a confirmation dialog, IMHO it shouldn't happen automatically in the first place. E.g., if executable code (e.g., plugins, drivers, codecs) need to be installed, let the user explicitly download and install them separately.
- more importantly: don't fscking nag. If the user explicitly refused something once (e.g., installing a certain plugin) or explicitly disabled something in the menus, remember that and _obey_. Don't hit the user with a barrage of popups asking again and again that maybe he still wants to reconsider.
E.g., if you disable ActiveX in IE, it will nag you with popups to re-enable it. E.g., if no, thank you very much, I don't want to install a certain plugin, IE will nag me to install it. (And worse yet, each time it will first hog the bandwidth downloading it and _then_ ask.)
I'll tell you a story.
/. about MS security, btw.) Joe Average wouldn't know, and wouldn't reformat.
I once had to install Windows 2000 on a box, and as Loki would have it, I had no Zone Alarm or Sygate Personal Firewall on a CD at hand. Just as Joe Average would.
So I could go download it somewhere else, or I could do a scapegoat installation just to download a firewall. I chose to just sacrifice an install to the gods of Hacking. I _knew_ I'd get hacked, but that was OK, since I'd reformat immediately after anyway. (Takes less time than whining on
(And I'm not disappointed. It takes less than a minute to get my uplink bandwidth saturated with mysterious outbound packets.)
Still, it will serve to illustrate what happens after you get your machine 0wn3d by some l337 skr1p7 kiddi3.
So I decide to play with it a bit longer, and see what happens with a firewall and an 0wn3d machine.
I start the newly downloaded and installed Sygate Personal Firewall, and immediately it pops up a window telling me the name of the application _and_ what's it trying to do. I block it, and that's that. No more outbound packets. I can tell struggles long and hard to send crap, but it can't. Both its inbound and outbound pipes have been sealed shut.
I can now toy with that machine as long as I wish, trying to disinfect it. Again, which is what Joe Average would want. If it's _not_ a sacrificial install, but some machine where his resume and a few gigs of other important data is, Joe will not want it reformatted.
I can even surf the net looking for information on the trojan, safe in the knowledge that it's blocked. No need to pull out the network cable.
Whereas you tell me that Apple would have allowed it to open its own ports, as it damn pleases. Inbound or outbound, whatever. And not even told me about it.
Well, gee. Sorry, that's not the kind of security I'm looking for. Dumbing down a firewall to the point where it doesn't actually block anything, in the name of "user-friendliness" is _not_ the way to go.
"1. All-in-one = single point of failure when something breaks, and needs to be replaced."
.mp3, the CD, the DVD, or the video game???
.mp3, the CD, the DVD, or the video game???" It plays whatever you've just put in that one drive. Just like any other console.
Bullshit. See the airplane principle: an airplane with two engines has twice the probability of an engine failure. _If_ I was to assume that electronics have a high failure rate, I'd be more worried about having 10 modules than about having one.
But the failure rate isn't that high in the first place. My PS2 still works flawlessly. For that matter, so does my old Playstation. So does the Dreamcast I used to carry around everywhere. (Man, I loved that thing.) Admittedly, they've been well treated, but then I'd assume most people who are serious about watching DVD on one don't go around pouring Coke in them either.
And if it was to break tomorrow, wth, I got years of gaming _and_ DVD playing out of it. Methinks that's great value for the money.
And if we're talking reliability, let's talk a very related point: the advantage in using standard, proven components. The only ones experiencing any significant failures were the ones who went their own non-standard, non-converging way. Sega even got bitten hard by that, as their first units had massive problems with the still experimental "GD-ROM" drive. The ones who just went with a standard format, like DVD, even for convergence reasons, actually ended up with a more robust product.
"2. Generally speaking, all-in-one devices incorporate propietary technologies to promote lock-in and/or reduce 3rd party tech licensing costs for the company (SONY!!!!)."
Bullshit. You don't need proprietary DVDs to play on a PS2. Any old DVD will work.
Au contraire, if you want "proprietary" and "lock in", look no further than Nintendo. Or Sega. Where Sony put a bog-standard DVD-ROM drive in the PS2, both Nintendo and Sega went and made their own obscure proprietary CD formats.
"3. Quality of stand alone components is usually much higher. Think stereo equipment."
Bullshit. Or at least missing the point. Not everyone is a rich consultant, who must have some $3000+ designer module even to play CDs.
Sony's components in the PS2 are perfectly on par with most consumer grade stuff, and actually better than the cheap on-board sound you'll find in most PCs. And you get all that for a mere $15 extra. For most people it's superb bang per buck.
"4. Modularity = more cost effective upgrade path."
Bullshit. More cost effective than $15 for the DVD playing capability? You must be smoking some really good shit.
"5. All-in-one = usually more complex than individual devices. Stand alone means you can learn and understand the functions fully before moving onto the next component. Sometimes the 'role' of a device is confused when it is consolidated. e.g. Does 'play' mean play the
Bullshit. Have you even seen Sony's remote control, or are you talking out of your ass?
Hint: It's a standard remote control, just like those you'd get with a standalone DVD or VHS player. Anyone who can't figure that out, is probably too stupid to use any other remote control too.
Plus have you even seen a console? It only has one drive. How the heck would anyone get confused about "Does 'play' mean play the
And it has auto-play, just like Windows. So you can just see it on the screen if a game started, or you get a DVD's menu.
Briefly: anyone stupid enough to have trouble with that, would probably be too stupid to figure out whether they're playing Donkey Kong or Castlevania on their Nintendo 64. Or for that matter whether they're playing Doom 3 or Solitaire on their PC. I.e., at that point the problem isn't the complexity, it's just a user with single digit IQ.
"6. All-in-one convergence not always a logical combination. Digital camera cell phones? mp3 playe
Guess one can never get enough of "waah! but what if someone had patented sex!" kind of fallacies on /. Was starting to get withdrawal syndromes after through a whole weekend without reading one ;)
But OK, let's play that game. Let's talk about paintings:
1. They'd more likely have to patent a device or method to make those paintings. So someone might have got a patent on something new like flinging colours at the canvas, but then someone else might just as well get the same result (or close enough) by using the old methods (using a brush). For which plenty of prior art existed.
(Just as this patent doesn't prevent you from having a lit case by other means than what Apple patented. You can still have your old cathodes or LED fans.)
2. For that matter, it might have stimulated someone to try more new stuff. So we might have 3-4 times more styles in the same period. Which is the whole purpose of patents: to stimulate researching _and_ publishing your research.
(And you could say the same about the situation at hand. We've had _years_ of noone even trying something more original than yet another LED fan or cathode behind acrylic window. By now every kiddie has one of those. So if it takes patents to get out of that loop and have a more original case, seems to me like a benefit of patents.)
3. Patents are not for ever. Copyrights amd trademarks do get extended. Patents expire no matter what.
I.e., if you talk about a 100 years interval, you may notice how the 20 years covered by a patent is only a fifth of it. I.e., combined with the previous point, we'd probably have a helluva lot more art choices after 100 years.
4. Patents encourage publishing your results, as opposed to keeping everything super-secret. Art is a bad example there. But there are a ton of technological processes that one could have kept secret. Or which _are_ being kept secret. Patents encourage companies to share this information with the rest of the world instead.
5. Patents get licensed all the time. I'm sure that if someone absolutely needed to do something impressionistic before the patent expired, they could have negotiated a license.
(And in this case, if IBM or MS absolutely needed to do their own lit cases, I'm sure they can persuade Apple.)
I don't know even about very young children. I'm getting the idea that the only ones who make "educational games" are the ones too fucking unskilled to make a proper video game.
Can games teach people stuff? Well, yes. I've learned a lot of history stuff from games like Europa Universalis, or to a lesser extent Civilization. Or at least got the curiosity to read more about that from other places.
Or "Die Gilde" ("Europa 1400 - The Guild" for you 'merkins) gives you a historical report of what happened IRL in that year after each game year. I've learned more late medieval trivia from it than from any other game.
But here's the scoop:
1. It must be fun as a _game_. Civilization was a bestseller in its own right. It didn't need to masquerade as "educational software" to get any sales at all. Ditto for Die Gilde, at least in Germany. Europa Universalis has a steep learning curve, but also got quite a few people addicted on its own merits.
2. Don't lecture or preach. It must first and foremost be a game, not a piss-poorly disguised beating people up with a clue stick. People instinctively resist being lectured.
3. Don't be patronising. Stuff that basically says, "see, we know you're a fscking retard who doesn't know how to put stuff in a fridge. We also know you're an idiot who can't figure out the cheapest crap to buy." serves no purpose other than humiliating someone. If anything, it'll make them resist the lecture even more.
And I'm thinking the same could be applied to software for small children. A game should be written to be first and foremost a _game_, and only incidentally also education.
E.g., there are a ton of _fun_ ways to make someone exercise their maths or logic skills. Economic sim games have done that for ages. Puzzles are also a good means to that end. (And god knows even the worst maths puzzle is still better than yet another "jump puzzle".)
So it's not like they _have_ to be crappy _and_ patronising games to be educational. It's just that the people making them seem to be into patronizing their gamers. And in most cases also utterly unable to make a proper game anyway.
Well, the thing is, the ads in magazines are very targetted. They fit the narrow scope of the magazine, and if I'm blowing some money to read that mag, chances are I'm interested in that topic.
E.g., if I'm reading a hardware magazine, chances are the ads will be hand-picked to be about hardware. If I'm reading a Java magazine, it'll probably have _something_ to do with Java. (Even if sometimes as little as "our new Ultra-Expensive Series (TM) servers run Java very very fast.") Etc.
It works on the Web too, not only in mags. E.g., I bought City of Heroes because several online comic strips ran, well, strips about it. They weren't intended as ads, but they worked as such admirably. It was as targetted as it gets: I was reading _comics_ about _games_, so it's a safe bet that both rate high on my interests list. So when they mentioned a game about comics matched both.
However, the thing is: they're hand-picked.
I don't think simple keyword matching can come even close to that. For web advertising to truly work, it would have to be a lot more targetted and relevant than it currently is. It needs a human to do the matching between ads and content (or at least the site's general direction.)
Well _if_ targetted marketting was actually smart, and not just word matching, that and probably a lifetime supply of game ads. Since mostly I'm on game related boards anyway.
Of course _really_ smart targetted marketting would offer me an auto-trolling program, so I don't have to type all that stuff. One Click Trolling. Someone ought to patent that.
Of course, being just keyword matching, I generally still get only irrelevant stuff.
For example I check out the lyrics to a gothic song about death and funerals. (Standard goth stuff, really;) And, presumably because it mentions "eternal sleep", I get an ad for sleeping pills. Erm. I really don't think they meant that ad to be targetted _that_ way. Just because I listen to gothic stuff, doesn't mean I want to commit suicide, dammit.
If you actually started recording surfing habits through these keyword matchers, it would get even more funny. For example, I'm sure that based on the words on those gaming boards, someone could conclude that my interests include:
- archery (what with all those bowmen/crossbowmen/etc in strategy games)
- horses (stable buildings in games, the occasional historical debate about stirrups and knights, etc)
- nazis (well, they're in every other game)
- voodoo (well, zombies anyway. They're a popular target in games) and generally magic (RPGs. Nuff said.)
- stealing (can't have an RPG without thieves)
Etc.
I shudder to think what kind of ads I'd get if someone actually collected that data.
What did you want me to do? Write a whole tome worth of everything that ever was wrong with Java client side? I did say "That's just a small slice of the many ways in which Sun started it on the wrong foot."
Believe it or not, I've been in client side application projects. I also have Eclipse open right now.
In fact, Eclipse is the prime example of how Sun started on the wrong foot. SWT (the widget set used by Eclipse) is IBM's work, while AWT and Swing are what Sun had in mind. The words "dumb and dumber" come to mind, when I think about AWT and Swing.
Want one thing that an actual client complained about? We had to program the same very small app in C _and_ Java, on Windows, Linux _and_ MacOS. Well, the whole package were more apps, but this was the only part which had to run on all desktops.
The client's first comment? "Whoa! Why does the Java version need 16 MB of RAM, when the C one is less than 1 MB?"
Another project was a bigger Swing "thick client" database program. In Swing. You know what the first complaint was? The look and feel. Inconsistent fonts, inconsistent keyboard shortcuts, etc. Sun's "Windows" Look and Feel is a sick joke. We ended up doing more work to write our own look and feel, just to get the program to look like a native app.
Yet another way in which SWT got it right, while Sun screwed up.
The second complaint? Performance and memory use. 'Nuff said.
Also, believe it or not, bigger companies have tried moving to Java on the desktop. It was every company's wet dream to suddenly only have to maintain one version, and work on every single OS. Some of which they didn't even support before. Everyone tried it.
E.g., Corel at one point wanted to move their whole office packet to Java.
Why do you think Microsoft got so scared and wanted to kill Java? Precisely because of this. It looked like all of a sudden everything will be multi-platform, and the "but this and that runs only under Windows" advantage will evaporate over night.
What happened to all those projects? They flopped. They ended up slow, bloated and more unstable than their Windows counterparts. So everyone moved back to C/C++.
If that doesn't count as a monumental flop, I don't know what does.
Doing at least basic image processing on photos taken with their digital camera? You'd be surprised how many moms and pops at least try their hand at photoshop nowadays.
Some form of media encoding? You don't have to be a genius to run a file through DivX, you know. Or to rip your collection of music CDs to MP3s, so you can put them on an iPod.
Antivirus scan kicking in? You haven't really learned what "slow" is, until you've witnessed Norton's or McAffee's scan kicking in. In fact, much of the popularity of, say, Intel's HyperThreading isn't because people have heavily-multithreaded apps, but because you can still reasonably use that computer while virus scans and such kick in.
(Not that their real time scanning is any better. Watch the same directory full of small files take under a minute to copy to a file server, without Norton Antivirus. Watch it take more than half a fscking hour with Norton Antivirus on.)
And so on and so forth.
And even games aren't that unheard of among the less computer-savvy people. Just because someone isn't a die-hard smack-talking rocket-jumping Doom 3 player, doesn't mean that everything they'll ever play is Solitaire.
Java is pretty popular on the server side, but client side it was always one monumental flop.
As applets go, for example, nowadays the whole program-inna-browser market is owned by Flash, followed by ActiveX. And for good reasons.
Starting with the fact that Java 1.0 was indeed a slow piece of crap for anything but the most trivial applets. Try displaying a complex table without a JIT, and you were talking about response times you could measure with a stopwatch, not with System.currentTimeMillis().
The initial lack of support for packing everything in a jar didn't help that cause either. Downloading 50 classes as separate files isn't particularly fast. And that's a very small project.
And for all the multi-platform hype, wasn't particularly portable either. If you tried running even a trivial AWT applet on different platforms, you wouldn't even get the same events. Or for something which required you to give a size in pixels on the web site, you wouldn't even get the same font sizes.
And by the time it caught up... meh. Flash is _still_ the better choice.
Not the least because of download size. Sun now includes all the crap they could think of as standard libraries. Do I need an XML parser to make a simple game applet? Not really, but Sun wants my users to download that crap anyway.
(No, it's not a made up problem. I've had modem users tell me literally "whoa, I'm on dialup. Is there some smaller version I can download?")
That's just a small slice of the many ways in which Sun started it on the wrong foot.
We have a thread which basically only says "let's try to attract them to CS. Who knows, they might even like it?"
We're _not_ talking about enforcing 50% quotas. We're _not_ talking about punishing people for not hiring women. We're _not_ talking about any extra incentives to hire women either.
It's _only_ about introducing some people to programming. Just a stupid summer camp with computers. That's all.
Yet someone feels a need to throw a tantrum about "feminazis" and "man-haters", and inherent gender differences, and how it's ok to keep them in underpaid underling roles. 'Cause, you see, those are important too.
Someone for whom the very idea of women even being invited to a stupid CS-themed summer camp, is already enough to trip a whole rant.
You're talking about conducting an interview to see who's got the skills. (Much as I believe objective CS interviews to be on par with Santa Claus and the Tooth Fairy, but at least the intention is there. Or the wishful thinking.) Whereas he is going on a tantrum at the mere idea of women being introduced to computers.
Sorry, that _is_ discrimination.
Dunno about you, but I'll call him a flaming idiot all right.
Dunno about you, but I hope such bigotted idiots aren't allowed anywhere _near_ a hiring position. Someone who's mortally offended by the idea of even letting women near a computer on a summer camp... there's no way in heck I can believe that such people can be objective and biased in a job interview. More likely he'll just go offended by the idea of women in non-underling roles again.
Look, I'm not demanding quotas or anything. (Although, come to think of it, it couldn't make hiring in this industry more of a joke than it already is, anyway.) I'm not saying to prefer hiring women or anything.
I'm just saying: start with an open mind, and at least try to keep prejudices out of the way. And FFS, don't throw tantrums to the mere idea of a CS summer camp for girls. That's all.
And if you're already doing that, well, you're not the kind I was calling an idiot anyway.
Well, where are the good male programmers? About 3 out of 4 can't program at all. They just don't have the intellectual capacity for it. More than 2 out of 3 don't even know the language they've put on their resume.
/. that passed that test. Invariably stuff like "oh, my biggest problem is that I'm totally focused and excel only at what the job description says". Only phrased in a way that snowed a clueless HR droid. Well, gee, anyone whose biggest problem is being a great professional, must be more perfect than Jesus.
/. pointing to people who are proud of being arbitrary just because they can. E.g., dropping CVs just because they don't like the email address. Or various other examples of an insecure incompetent having power trips along the lines of "fsck you buddy. I don't need you. You're the peon grovelling for mercy at my feet. So let's see you convince me why should I waste my royal time and bestow the undeserved favour of a job upon you."
I work with some people which don't even understand what "call by value" means, or what a hash table is. At all. And some competent people, but more of a coincidence than any merit of whoever did the hiring.
So I'll go further and say: where are all those interviews which actually determine someone's skill? I've yet to see any job interview which was even a test of aptitude. At all. It invariably was a "prove to me that you're willing to bullshit your employer with a straight face" contest.
Being a honest guy, and having enough real stuff on my CV anyway, I say just that. On the other hand, invariably I can tell that I could just as well have said that I was the IT Grand Vizier to the Sultan of Cyberia, and would have got the job just as well. And invariably in the same week someone completely incompetent got hired as well.
Often complete with bullshit questions like "what is your biggest deffect?" No, it never ends up testing someone's honesty. It's just an exercise in creative bullshitting.
Look even at some examples cited on
Or we've had article right on the front page of
I've yet to be asked some relevant question. Like, say, "when would you _not_ use <currently hyped and fashionable technology>". (Notice the "not" there. Any monkey can learn the hype, but it takes some skill to do actual design.)
So where are all the interviews which objectively prove who's the better professional? No, really. I want to know. Maybe I'll apply for a job there next time.
So for me to believe that a job interview objectively proved some gender/race/nationality/whatever differences, I'd first have to believe that it objectively proved _anything_. As long as it's mostly just a "which of these guys I like the most" contest, it'll at best just be an exercise in reinforcing a pre-existing prejudice.
Disclosure: I'm male. I also believe that respect ought to be awarded for _personal_ merits, not by just being born into the "right" group.
And basically what you're telling me is along the lines of "underpaid underlings are signifficant too, so it's perfectly ok to force women and minorities into those roles. And therefore it's ok to pay them less."
Gee, what a flaming idiocy.
You know what? It's not the PC crowd that are the illogical zealots, it's the relics from the past like you. Idiots who think they should be paid more just because they happened to be born white or male or whatever.
Well, let me break some grim news to you: if your _only_ merit is that Y chromosome, you're a fucking waste of air. That chromosome alone doesn't even make you necessarily smart enough to flip burgers at McDonald. Much less make you qualified for science or engineering. Much less for CS.
If you want any respect or a better paid job, show me what _you_ personally have done to deserve it. Not what random thing you've got at the chromosome roulette.
And if you actually need to blame your problems on women, blacks, jews, corporations, international conspiracies or whatever, you've just earned a hefty dose of disrespect from me. Means you're not only an incompetent piece of waste, but unable to take responsibility for your own failures either.
Let me spell it out: anyone who's anywhere _near_ competent in their job or life, doesn't need a scapegoat to put the blame on. Sure, bring in the women, blacks, indians, or the mongolian hordes for that matter. I know I won't lose my job in the process. Or will find an equally good one quite easily.
Do _you_ feel threatened by "feminazis"? Excellent.
Consider yourself flamed.
When I was in the teens, I also had the idea to make a virus. Mind you, back then it meant file or boot sector virus, not an exploit over TCP/IP. We're talking the 80s.
Coding one was actually a very trivial exercise, and I like to think I did a fairly good assembly job of keeping it under 512 bytes, so it wouldn't be too obvious. It also worked.
But here's the catch: even at that kind of age, I was perfectly able to know it was wrong. That virus was only tested and demonstrated to a couple of friends in a controlled non-networked environment, then destroyed. It also carried no malicious payload anyway.
Teenager or not, I think everyone I knew was perfectly able to distinguish the moral difference betweem a "let's see how this works" exercise, and releasing a plague. Some of them might have thought it would be way cool to cause that harm, but I don't think anyone was incapable of discerning that it was wrong.
And let's forget viruses. We trust teenagers with a lot of other stuff.
E.g., depending on the country, you may be allowed to drive at 16 years old. You are trusted to drive a dangerous ton of metal, and not suddenly go running people over.
E.g., in a lot of countries you can be conscripted at 18 years old. And then given an assault rifle and live ammo, and left alone with it for hours, guarding some corner of the compound. Someone incapable of discerning consequences would start shooting at people to pass the time away. But in practice they don't.
So, well, gimme a break. If teenagers can be trusted with an assault rifle or a car, I do believe that an 18 year old is also very capable of understanding what a virus is.
I'm such a geek. Normally I don't really give a damn about what people think of what I wear.
(It's decent. I'm not wearing a crotchless S&M outfit to work, or anything. If anyone is fundamentally offended the sight of a clean pair of jeans, they're just stupid. And I have better things to do than worry about stupid people.)
However, in this case we're talking a court of law. You don't want to piss off the judge who might, on a whim, give you a suspended sentence or community service or send you behind bars for a few years.
You _don't_ want to look like an unrepentant "fuck you all" rebel to the judge. You don't want to look like you're damn proud of what you've done. (Which is the impression that such a "Big Daddy" t-shirt would have given even me.)
I'm not even saying he should have worn a suit and tie or anything. But, you know, even if you're gonna wear a t-shirt, make it a plain one.
I mean, geeze, wear that t-shirt to school. Wear it at a party. Wear it even to a job interview if you honestly don't give a damn about the outcome. But a court of law is more serious: unlike a job interview, you can't just try again somewhere else.
Basically all I'm saying is that there's difference between not caring about stupid people, and _being_ the stupid one. Freakin' big difference.
Actually, while QWERTY wasn't designed to slow people down, it _is_ designed to avoid jams.
The thing is, the contraption consisted of (more or less) a semi-circle of thin levers, each with a little hammer with an embossed letter on it. All were aimed at the same position on the paper. You press a key, and purely mechanically the lever would swing the hammer at the paper. (Well, actually, at the ribbon.)
Also, because it was a purely mechanical contraption, the cheapest and most reliable way to build one was: keys that are close on the keyboard, would also activate levers which were close to each other.
Jams would happen when two close enough levers would be activated at the same time. Or close enough. The closer the levers were, the more likely you'd get a jam. (Again, purely coincidentally, this also meant "the closer two keys were".)
E.g., pressing "Q" and "P" at (almost) the same time would never jam. They swung from opposite directions, and it was pretty much guaranteed that one hammer would simply hit on top of the other. E.g., "A" and "S" at the same time (e.g., while typing "ASSASSIN") would pretty much always jam.
So basically, QWERTY:
1. was just supposed to prevent jams. (Which cost more in typing speed than a couple ms worth of more finger movement.)
2. was not designed to do anything to typing speed as such. Neither maximize it, nor minimize it. Whatever typing speed difference it produced, it was "side effect", rather than "goal". (And, again, a lot of it came from jam prevention rather than anything else.)
3. the _only_ typing speed consideration it received at all, was a rigged tech demo. Ever wondered why the "QWERTYUIOP" row? Because the rigged tech demo was basically "Look! I can type 'TYPEWRITER' quickly! It must be an optimal layout!" Hence all the letters in the word TYPEWRITER had to be on a single row.
(Hardly a scientific study, but PHBs bought it anyway.)
Furthermore, I'd point out that:
A. It did a piss-poor job even at spacing common letter combinations apart. E.g., even in their tech-demo "TYPEWRITER" they have letters which are near each other: "TY", "EW", "ER", and thus prone to jamming. "W" and "R" aren't that far apart to be jam-proof either.
B. if you've ever used one of those purely mechanical typewriters (no, some electronic thing doesn't count), you'll notice that typing was a different exercise on those. It involved keeping your hands above the keyboard and hitting the keys pretty hard. At the very least it's _not_ the same RSI prone position you'd use on a normal PC keyboard.
C. a PC keyboard doesn't jam.
D. Even if you do type the wrong letters on the PC, the cost of errors is next to nil. Correcting a mistake was a _very_ time consuming operation on a mechanical typewriter, since it involved physically erasing or covering printed stuff with white paint. By comparison, hitting backspace on the keyboard costs a small fraction of a second.
Etc.
So basically I'm saying that the considerations from which QWERTY was born, not only were imperfect to start with, they bear exactly _zero_ relevance to a computer keyboard. That QWERTY still works well, is more of a testimony to the fact that people can learn _any_ keyboard layout well enough, than some inherent advantage.
QWERTY, Dvorak, even alphabetical order, IMHO you probably just type faster on whatever you have more exercise. That's all.
The thing is, far from Apple having a revolutionary new model, they're just stuck at a model that existed ever since the first computers.
In fact, in the beginning, there was no software market at all. When you got a computer, say, from IBM, the _only_ software you could ever hope to get was also from IBM. That included the OS, assembler, compilers, utilities, _everything_.
And chances are you couldn't plug in any peripherals that your vendors didn't sell, either. Some vendors went as far as to patent the connectors, to keep other people from undercutting the price of their ridiculously high priced peripherals.
That kind of a lock-in isn't new, and isn't to keep the customer happy: it's most marketroids' wet dream come true. It's pretty much like having a captive market, sorta like the 17'th century colonies. Or like having serfs paying the yearly Sun/IBM/DEC/whatever tax.
Except this time the fear was not of the King's army, but of the costs of rewriting everything to a new platform.
Even when Unix came and (theoretically) gave the proprietary OSs a kick in the pants, ever wondered why the Unix fragmentation happened? (Which in turn paved the way for Microsoft's defeating them all.) No, seriously. Because noone actually wanted to have a compatible platform, which would break their customers free from the vendor lock-in.
A truly portable Unix and tools, would have allowed you to just take your programs off that DEC mini and move them to a Sun. Or viceversa. Pretty much forcing a competition strictly on price/performance, instead of a cost artifficially skewed by the migration costs.
So what did everyone do? Made their Unix version slightly different. Made sure that although theoretically they were all Unix and all ANSI C, you couldn't really just take your programs and scripts and move to a rival's computer.
It's not even limited to whole computers: see how 3DFX guarded Glide jealously until right before the end. They wanted people locked into needing a 3DFX card, instead of free to get anything.
Or not even limited to computers at all: see, for example, how Apple doesn't want Real being able to sell tracks that can be played on the iPod. They want you locked into having only one vendor of DRM-ed music for your iPod.
So basically, again, Apple isn't really doing anything new. It's just doing the same old crap of trying to lock the customer in.
And if that's the great liberation from Microsoft's tyranny... well, count me out. I'll stay here where at least I still have _some_ choice. I can at least understand the Linux crowd to that end, since Linux _is_ free of proprietary crap and vendor lock-ins. But Apple? Gimme a break. They're twice as bad nowadays as MS and IBM put together ever were.
Anything worth protecting in a company, they'll patent. You know why? Because patents give you a 20 year unconditional monopoly, while "trade secrets" in a product you can just open up and copy are a toss anyway.
.gif roll-overs per page."
Trade secrets are valuable when you can actually keep them a secret. E.g., some secret stock picking algorithm that runs only on your computer, which isn't even talking directly to the web server. They just share a database server, past 5 firewalls and 2 armoured doors. Someone would have to actually break into the building (or bribe the sysadmin) to get to that.
But in a product which you ship and anyone can take apart and/or reverse engineer? Gimme a break. There's a reason why even software corporations want patents instead: because they're a far safer bet than pretending it's top secret.
The way Seagate's GMR heads work, is not only the same as everyone else's work, but you could just take the drive apart and find out for yourself anyway. The way their caching and TCQ algorithms work, you could just disassemble the firmware and see for yourself.
So IMHO Seagate is just having a knee-jerk reaction, whether they have anything to protect or not. They're just doing it because everyone else is doing it, not because someone sat down and analyzed what secrets they have to protect.
Same as we've seen during the dot-com scam. People made you sign NDAs to protect some super-secret plan like "umm... we're making a web site. With a forum. And... umm... yeah, I know, lots of Flash, bright colours and 1 meg of
Well, geee... surely noone else would have thought of that on their own. Noone except the other tens of thousands of equally clueless dot-coms, that is. And surely noone can get that idea by just seeing the bloody web page. They surely _have_ to steal your employees to figure that out.
Yet they made you sign tome sized NDAs. One I had in front of me wanted to prohibit me from ever competing not only with them, but also with their clients, or their clients' clients' clients. In any way. Taken literally, if they ever made some ad for a news site, I'd have been prohibited to ever blog, 'cause that would sorta be competing. If one of their clients was a detergents corporation, I'd have been quite literally prohibited from selling soap.
And, again, to protect what great trade secret? That they want to make a web site. With flash. And ads. Gee, that's so revolutionary, non-obvious and worth protecting.
So personally I'd very much like to see this idiotic cult of the NDA die already. My message to these employers would be: if you have something worth defending, get a patent already. If it wasn't worth patenting, go fsck yourself and get off my case already.
I don't think that structure or strong typing precludes good design or tests. You can have it all.
Tests are good and fine, and everyone ought to use them. Indeed.
But keeping me from passing the wrong arguments to the wrong function is another source of bugs eliminated. A source of bugs which can sometimes be rather insidious, since parameters can come from anywhere and get passed through 20 function calls. They can come from a user who (sometimes even deliberately) enters wrong data, the web designer who did this small change in the template and replaced the numeric ids with string ones, or the poor maintenance programmer who has a deadline to fix something he doesn't even understand.
And let me stress the last part: maintenance is the bigger problem, not writing the code. It's also the under-manned part: while when coding you could get only 10,000 lines per person, in maintenance one single person can well get a 100,000 or even 1,000,000 line project alone.
That person sees the program like through a keyhole. It's like seeing the Sixtine Chapel fresco for the first time... through a pinhole. You can scroll the view, but you see a couple of square inches at any given time. Good luck seeing the big picture. Good luck remembering it.
But when you're doing maintenance you're not always even allowed the luxury of getting the big picture. You're just expected to fix this one small problem, and fix it yesterday. Just quickly find the line that does that, and change it already.
Which brings problems like:
- where the heck _is_ the code that does that?
- where does it get its parameters from?
- are those parameters what I think they are? (E.g., does it always get a ClientDO object? Can it be null? Can it get some other data type instead?)
- I see FunctionX and FunctionY being called. Which in turn call functions A, B, C, D _and_ E. What _are_ their parameters? Am I sending the right data?
(E.g., I see it expects a Collection object. Will _any_ collection do? How do I know it doesn't get cast to ArrayList down the line. Don't laugh, I've seen code like that.)
- is my change breaking anything else? (E.g., passing the wrong data type to a function.)
And at that point _any_ help is most welcome. Good structure and strong typing and unit tests can give that maintenance programmer vital hints. Hints which otherwise he/she would have to spend weeks digging in code for.
That's really when that stuff is the most valuable. Sure, when coding you can often do without them. It's maintaining someone else's code that's the real kick in the nuts.
Stuff like Heisenberg or Einstein's relativity are, no offense intended, most quoted as "science doesn't have all the answers" by people who don't understand them at all.
Heisenberg itself doesn't just say "buggrit, you just can't know where the electron is", it gives you a formula there. You can actually calculate stuff and build practical stuff with that knowledge.
(E.g., the Zenner diode. E.g., Heisenberg may well be the reason why CMOS miniaturization comes eventually to a halt. And it offers you the formulas too to know what voltage you need to keep enough electrons there, for a given transistor size and shape. Etc.)
Second, it too can be observed experimentally. It's not used as way to throw up your hands and say "the electron does whatever God wants", but to understand and _predict_ where the electron can go. And within what margin of error.
Is there a ton of stuff we don't know about yet? Most likely. Are there limits to our theories than we don't know about yet? Almost undoubtedly.
But we'll cross that bridge when we come to it. And then we'll have an even more complete theory, and more advanced experiments to test its limits.
And in the process we'll be a lot more advanced and able to build more useful stuff, than if we had all just thrown our hands up and said "buggrit, it all happens because God wants it to." Well, sure, that may be so, but God's will alone hasn't produced any predictable results so far: you can't build a transistor based on it