With the exception that you then have to burn one cd at a time and at a cost of around $.25 for a good disc. Disc printing costs are down to the pennies a disc level and you can have them all pressed at the same time, with a kickass screen print on top basically "for free."
Generally bands that record on CDR are considered small potatoes (or perfectionists), as it's cheaper when doing 100 discs or more to have them pressed.
Tons of people in the computer lab I used to work in would keep AOL floppys to save stuff on, because even though they were totally unreliable they were abundant and free.
This would be a really good idea for bands jsut starting out. Record a CD with three songs and leave the rest blank, give away free. People burn other stuff on the end, and hear your tracks first. Free advertising marketed to people who might actually dig your sound.
Of course, you could do the same with recorded commercials...
This is basically a procedural vs. modular question. Scripting languages are generally perceived as being great for doing one task but useless for doing multiple ones.
That is of course bullshit -- Perl, Python and WSH may have messy syntaxes (IMO) for performing OOP operations but they're obviously very good at it, and in some cases can be faster than "high level" languages. But I still prefer to use a more structured, accepted language to code things. Why? The IDE! If you're writing a massive application, it's great to be able to see things graphically. Scripting language IDEs (mostly because of the stigma against them) tend to be "syntax-highlight only."
Part of this is also the "right tool for the right job" mentality. Since scripting languages are used for automating daily tasks, many managers assume they're not up to par for designing larger, more critical systems. Once again, this is bullshit (i'd trust Perl for speed, reliability and fault tolerance over VB 6 any day), but the stigma exists. After all, if you say "I can do that 1 week app in a day," it sounds like you're cutting corners somewhere. Nobody realises the corners you're cutting are mostly syntactical "accounting" overhead.
I think scripting languages are gaining in popularity though. Notice the inclusion of regular expressions in Java 1.4 and.NET, and script execution classes for.NET. The leader here of course is the web. With the need to change things quickly without the need for instant gratification, suddenly scripts made MORE sense than compiled code.
Re:Zire is the only one that make sense
on
Palm PDA Roundup
·
· Score: 1
Ahhh here it is...Missing Sync, http://www.markspace.com/pocketpcpreview.html
I think I trust it more since they already have a version for the Clio, that's apparently pretty good and fairly priced. As opposed to charging a pantload for a beta:).
Re:Zire is the only one that make sense
on
Palm PDA Roundup
·
· Score: 1
Pocketmac is beta software. It's expensive and from what I hear it doesn't work very well. Basically, you're paying to test for him and there's hints that he's going to charge for upgrades at some point.
You know it's not easy to do what he's doing, he's going alone and against the grain of both companies, but I'd rather just rely on my own means to get data to and from the palmtop.
There's another product...can't remember the name off the top of my head...that's taking a different stance than pocketmac in making the two work together. Rather than make a mac product that pretends to be ActiveSync, these guys are making the pocketpc work like a mac "native" device, adding support for it in iTunes and syncing it with iCal/Rendezvous. Personally, this sounds like a much better route...plus it's much cheaper ($35 vs $50/$100 for pocketmac).
Of course, this is all assuming it works.
Incidentally, if you're a mac/pocket pc developer interested in helping with an OSS project to connect pocketpc to the mac as a removable drive, thus eliminating the middleman USB card reader i'm using now, let me know. I've been thinking about starting one.
Re:Zire is the only one that make sense
on
Palm PDA Roundup
·
· Score: 1
I'm a mac user and I love my Pocket PC. Although I should point that unless you are fairly proficient with both devices your goose is pretty well roasted to get them to interoperate.
But since all I ever do is xfer mp3s and DivX files to my CF card, it works fabulous.
How secure can a number that can't be more than 4 digits long BE?
I hate my pin number. I hate that they won't let me use a longer number like I want to. Jesus, I know Pi out to 50 digits...yet the number that exposes all of my funds (yes, they forced me to have the same pin to link my savings, line of credit and money market accounts) is like the combination to a moron's luggage.
Hey you ATM hackers. How much work would it take to make these goddamn things accept a 12 digit pin -- or better still, a passcode.
And to whomever's about to bring up biometrics: shut up.
Computers already realise this and it's why they haven't taken over. At least, that's what my t-bird told me shortly before it over heated.
"What's the fucking point, meat man?" it said to me in its best Carl impression.
This is why all new IBM processors are coming with pseudoreligion sub processors. Get yourself one of the BUD-H1ST models, they work the hardest and don't try and start shit with incompatible neighboring chipsets.
Especially since the whole concept of "Singularity" is central to this fairy story. Singularity is a foolish term if you ask me...since fans of this bullshit like to refer to themselves as "theorists" and thus butt heads with those of use interested in real singularities, e.g. points of finite mass with no measurable volume found in the center of black holes.
Singularity "theory" presupposes a lot of sci-fi hocus pocus about machines being instantly better than us at everything we do well, including the reduction of their own power consumption needs. It's crap and requires such a heavy suspension of disbelief on the part of the reader that it's on about the same level as cypherpunk fiction in the 1970s or giant robot cartoons. With the probably exception that I LIKE cypherpunk and bigass robots...Singularity is just paranoid screwiness.
Oh get off the high horse. Just because something is deprecated does not make it wrong. HTML v3.2 is going to last us a long, long time because it is a) ubiquitous b) easy to understand c) easy to implement. It is becoming the HTML people "learn in school," the vulgate language.
And you know what? It doesn't hurt anything. The page downloads fast enough, and it renders nicely. Slashdot is an informative, easy to use web site, not a pretty one. And that's the way we likes it.
(oh, and i think the style strip is due to some ancient browser vulnerability that was used to make goatse.cx pop up. you forget that just because you don't use the cool new features of CSS to do ungodly ugly things doesn't mean others won't. and CSS is still so damned complicated and varied by implementation that it's probably more beneficial to stip it out than it is to learn how to clean it.)
Wow. I wish I would have thought of that before I diss'd them on slashdot...but that would require a modicum of faith in the cultish and often divisive open source community.
Oh, and i modded myself down on this post because i'm a troll.
Besides which, you can't compare a VIA C2 to a G4. Hell, even comparing it to a G3 is sort of like comparing a scooter to a Harley. Of COURSE it's half the price...it's less than half the machine!
And it begs the question: why are we supposed to trust a company that can't even get it's own marketing straight? They're just a company with a name that's a bad pun on the most effectively marketed operating system ever...I'd put more stock in AmigaOS. At least Amiga is trying it's own thing.
It's not America, but human nature that's "consumerist." Haven't you heard some of the horror stories of child prostitution in Asia and Africa? The latter has a rampant problem with grown men raping young girls because of a common myth that sex with a virgin cures AIDS.
But of course, that's only what I read in the news, and I can't imagine African news is any more interested in truth and level headedness than our American FOX/CNN/MSNBC cloud
The whole reason child porn is illegal is the common idea that children are easily misled. I'd like to say, as I did in high school, that I disagree and think that kids, teenagers especially, are capable of making their own decisions. But now I disagree.
Here't the thing: just having a naked picture of a child isn't going to incite anybody to do anything against their nature. It isn't the catalyst to the downfall of society and isn't going to create some massive underground rape culture.
But if it were legal -- as in legal to sell -- it would create an "industry." And because of the social stigma against child porn, it's not going to be a friendly industry. It's going to be made up of the worst dregs of the porn industry, full of trickery and lots of children would fall into it. Promised low incomes that sound fantastic to young ears they'll be easily pressured into doing things they don't want to do by people who don't care for them.
Making it legal to posess child pornography is like creating demand for a grey market in traffic and honestly making it worse for kids.
Do I think that pictures along are going to hurt society? No, and therefore I think sting operations and such are worthless as anything more than a deterrant. But come on. The world is dangerous enough with kids preying on each other sexually. We don't need to throw clever pervs into the mix as well.
Goddamn it, a picture of a baby in a bathtub IS child porn! I don't know where you parents get the idea that just because you find tiny naked kid parts cute that makes it okay to take pictures of them. Isn't it wrong to take naked pictures of a naked teenager? Cut the double standard. Naked baby pictures are sickening. I know I'm not the only person who things so.
Anne Geddes should go to prison for objectifying naked children. She's worse than Maplethorpe and Saudek combined.
And of course, if there is a list of sites to block, discerning porn fans from other states will get ahold of it and cut down their search times. What to do with the extra time? Why, trade on IRC. Censorship never works on the internet, it's way too wily
I read the British Columbia stopped it's pedophilia stings and pumped the money into an awareness program. I wonder how they're doing with that.
Case in point: Tom Pabst. The guy was the BEST reviewer of all forms of hardware in 1999. I wouldn't trust anybody else. The guy had the know how and the connections.
Then, he gave a couple bad reviews. To people like Intel. Suddenly, he wasn't getting the latest chips. Suddenly, he was borrowing them from other people, violating NDAs, or waiting for the release like the rest of us. His site became "great, but old news" instead of "just in time answers." I moved my eyes to the more "read between the lines" style of Anandtech.
Moral of the story? If you're a corporate whore, you'll benefit from it pretty much every time.
By the way, I love NT. Windows 2000 is the best system I've ever used and it's mostly because it developed organically. If linux had a bunch of eggheads worrying about ease of administration and intuitive GUIs instead of hooking everything through the command line and scoffing at developers who want to get paid, I wouldn't be tapping this into Internet Explorer.
Open systems have their own pitfalls though. They are far, far worse when it comes to bare bones implementations of hardware drivers because there really is no impetus to build a system that isn't immediately useful. Say what you will about windows, it's release quality. The parts of the API that don't work are on the fringes, new development, the same sort of thing that open system developers would change, too, if they didn't work the right way. Case in point: Apache 2, Cocoon 2, the infinite iterations of Freenet which are incompatible with previous nodes.
In the case of an open system, if I need to send data out of a widget, I build a driver to send data out. I don't spend any time at all on anything else. Another person who wants to suck data into a widget will have to edit the driver. There is no call for them to maintain any sort of version compatibility with mine; frankly, if they feel it's beneficial for their use to rape the core, they'll do it without any qualms. On the other hand, manufacturers of closed systems have to just that. They have to anticipate any way a person would use their driver, or else the hardware won't sell. Just ask Iomega about their Buz box, a device which never sold because they never released a 32 bit driver for it.
And in the end, since only one party's been touching the driver from a closed system, it's much cleaner and much more likely to have few points of failure. Fixing it is much faster because it's somebody's job to maintain that code. You don't have to download a source package touched by a hundred hands for a hundred reasosn and try to piece together what they each were trying to do. Of course, you have to rely on the vendor, but if you're an embedded systems developer you better damn well have a good relationship with your vendor as it is.
So the decision is thus: On one hand, you have open software. Where the freedom of use and lack of liability mean the possibility of a hundred apis broken off the same core with minor differences in the feature set which are all totally incompatible (PocketLinux). Where upgrading your webserver can break every other application on your machine (Apache) due to subtle dependencies for features you probably won't use and don't have the time to recompile for. Where continued improvement and a staunch refusal to freeze development so you can use it as an "API" forces you to maintain a constant watch on all new versions of all dependent code (GNOME, GTK, even glib). A platform where all of this work is placed on me, the developer, since it's too complicated for an IT person in our target market.
On the other hand, you have "POS" windows. A platform for which I can write a simple client server application in a day and have it run on millions of PCs without recompilation, and where most bugs are due to using the API in a way that isn't intended or necessary.
OSS rocks, but at its worst it's a maddening cob job. I like having the source when it's good, but there's a LOT of confusing shit as well. Better to put the responsibility for fixing a driver or api call on somebody who is both libel to me and who has experience in doing so.
The "Black Box" design theory abounds because of the freedom it offers programmers from the dark ages of having to know the underlying hardware intimately before anything could be accomplished. It's what allows programmers to devote all of their time to doing what matters rather than pouring over volumes of errata and arcana.
The reason Windows became so popular, for example, is because its API offered programmers a way to manipulate graphics without having to make graphics calls. Variation from driver to driver was of no concern, and shouldn't be -- that's an IT issue which can be repaired without redoing the entire application.
And in a perfect world, there's no problem. If a driver hooks into an API properly and documents any disperity, then the black box theory holds true. Problem is, driver aren't perfect. A lot of them are designed for bare bones functionality, and only optimized as necessary (hence how Nvidia's still squeezing substantial horsepower out of my ancient GeForce GTS with every new driver release). Obscure hardware cases always cause trouble, which is why Dells are (sometimes) more reliable than "no name" machines with "better" hardware. Dell has the clout to make sure the drivers are as seamless as possible.
What's the solution for embedded developers? Design and test the drivers in house, so the black box coders have a shoulder to cry on when hardware doesn't act properly. But it should not be the core developer's job to know what goes on with the hardware. That kind of thinking bloats budgets, increases the complexity of the project and ultimately the cost. Modularity, even though it makes things more difficult to map in total, makes things easier to deal with on a micro level. If the application works when unit tested but fails on the release machine, then it's the driver's fault. Much easier to fix than it is to perfectly replicate the release in your tests.
Expecting EVERY software developer to be an electrical engineer as well is absurd unless you intend to pay them for both degrees. Better to keep it modular and put the pressure on the hardware abstractor to do a good job of catching the tiger's tail.
Of course GOOGLE isn't buying it. Google's delivering nothing but text and small images. The technical bottleneck in everything Google does is bus speed and the internet itself. Logical solution? Parellelize the heck out of the problem and bollocks the "high tech."
Guy I know. His father's company just bought a packed X-serve RAID. They don't keep a lot of records and they can't go paperless due to restrictions in their industry. So they essentially have purchased a 2.52 TB MP3 server. It'll almost fit all his Zappa bootlegs.
At the same time, there are LOTS of customers out there who need 64 bit. Lots of folks who need faster and more robust searching. Customers who are dealing with sluggish behemoths from IBM that the Itanium (or, I prefer, the Hammer) is perfect for. Aerospace engineers who need to perform simulations and stress tests. And regular cats who just want incredibly high resolution math for doing speech to text, photoshop or gaming.
But GOOGLE? Google could operate on pocket calculators if you got enough of them, yahoo too. The web is wonderfully low-tech and low-resource if properly designed.
I wish more companies would follow their lead and use whatever technology is best for their model rather than something big, flashy and ultimately too expensive. This would lead to higher margins. What? Yeah, if you have fewer people clamoring for the TOL, you can work slower and charge more. Think Apple. Think Porsche. Then there's less of a need to farm everything overseas and release a new clock speed every month that's.01 GHz faster than the last.
I've been told that my credit is bad because of the number of times i've been denied refinancing loans on my car.
I try for these OFTEN because i'm two years into ownership on a car at 13%, pretty low at the time. I have excellent credit besides this (own a house, missed one payment in 7 years, high income to debt ratio).
Refinancing a car isn't like refinancing a home...a lot of banks don't want you to refi if a car is too old, because there's so much more risk involved. After all, if I defaulted and they repo'd the thing i'd still owe something on the order of 4k they'd never see. That 13% then is fucking me...unless i overpay, which i'm not willing to do, the point at which the car is worth more than the outstanding payoff remains in the distance, somewhere around summer of 2004. So there's all this risk, even with my bad credit, and no real reason for the bank to accept considering how little they'll make off the deal. And each rejection leads to another reason to reject me.
I'm not great with money but i'm also not a derelict. If only equifax would understand this!
Slashdot Mirror
With the exception that you then have to burn one cd at a time and at a cost of around $.25 for a good disc. Disc printing costs are down to the pennies a disc level and you can have them all pressed at the same time, with a kickass screen print on top basically "for free."
Generally bands that record on CDR are considered small potatoes (or perfectionists), as it's cheaper when doing 100 discs or more to have them pressed.
Tons of people in the computer lab I used to work in would keep AOL floppys to save stuff on, because even though they were totally unreliable they were abundant and free.
This would be a really good idea for bands jsut starting out. Record a CD with three songs and leave the rest blank, give away free. People burn other stuff on the end, and hear your tracks first. Free advertising marketed to people who might actually dig your sound.
Of course, you could do the same with recorded commercials...
This is also the way to solve the problem of "getting any work done."
This is basically a procedural vs. modular question. Scripting languages are generally perceived as being great for doing one task but useless for doing multiple ones.
.NET, and script execution classes for .NET. The leader here of course is the web. With the need to change things quickly without the need for instant gratification, suddenly scripts made MORE sense than compiled code.
That is of course bullshit -- Perl, Python and WSH may have messy syntaxes (IMO) for performing OOP operations but they're obviously very good at it, and in some cases can be faster than "high level" languages. But I still prefer to use a more structured, accepted language to code things. Why? The IDE! If you're writing a massive application, it's great to be able to see things graphically. Scripting language IDEs (mostly because of the stigma against them) tend to be "syntax-highlight only."
Part of this is also the "right tool for the right job" mentality. Since scripting languages are used for automating daily tasks, many managers assume they're not up to par for designing larger, more critical systems. Once again, this is bullshit (i'd trust Perl for speed, reliability and fault tolerance over VB 6 any day), but the stigma exists. After all, if you say "I can do that 1 week app in a day," it sounds like you're cutting corners somewhere. Nobody realises the corners you're cutting are mostly syntactical "accounting" overhead.
I think scripting languages are gaining in popularity though. Notice the inclusion of regular expressions in Java 1.4 and
Ahhh here it is...Missing Sync, http://www.markspace.com/pocketpcpreview.html
:).
I think I trust it more since they already have a version for the Clio, that's apparently pretty good and fairly priced. As opposed to charging a pantload for a beta
Pocketmac is beta software. It's expensive and from what I hear it doesn't work very well. Basically, you're paying to test for him and there's hints that he's going to charge for upgrades at some point.
You know it's not easy to do what he's doing, he's going alone and against the grain of both companies, but I'd rather just rely on my own means to get data to and from the palmtop.
There's another product...can't remember the name off the top of my head...that's taking a different stance than pocketmac in making the two work together. Rather than make a mac product that pretends to be ActiveSync, these guys are making the pocketpc work like a mac "native" device, adding support for it in iTunes and syncing it with iCal/Rendezvous. Personally, this sounds like a much better route...plus it's much cheaper ($35 vs $50/$100 for pocketmac).
Of course, this is all assuming it works.
Incidentally, if you're a mac/pocket pc developer interested in helping with an OSS project to connect pocketpc to the mac as a removable drive, thus eliminating the middleman USB card reader i'm using now, let me know. I've been thinking about starting one.
I'm a mac user and I love my Pocket PC. Although I should point that unless you are fairly proficient with both devices your goose is pretty well roasted to get them to interoperate.
But since all I ever do is xfer mp3s and DivX files to my CF card, it works fabulous.
How secure can a number that can't be more than 4 digits long BE?
I hate my pin number. I hate that they won't let me use a longer number like I want to. Jesus, I know Pi out to 50 digits...yet the number that exposes all of my funds (yes, they forced me to have the same pin to link my savings, line of credit and money market accounts) is like the combination to a moron's luggage.
Hey you ATM hackers. How much work would it take to make these goddamn things accept a 12 digit pin -- or better still, a passcode.
And to whomever's about to bring up biometrics: shut up.
Nihilism is pointless.
If there's no point to life, then there's no point to death. Which makes suicide is sort of presumptuous.
Me? I'm in it for the sex and the fatty foods.
Computers already realise this and it's why they haven't taken over. At least, that's what my t-bird told me shortly before it over heated.
"What's the fucking point, meat man?" it said to me in its best Carl impression.
This is why all new IBM processors are coming with pseudoreligion sub processors. Get yourself one of the BUD-H1ST models, they work the hardest and don't try and start shit with incompatible neighboring chipsets.
Especially since the whole concept of "Singularity" is central to this fairy story. Singularity is a foolish term if you ask me...since fans of this bullshit like to refer to themselves as "theorists" and thus butt heads with those of use interested in real singularities, e.g. points of finite mass with no measurable volume found in the center of black holes.
Singularity "theory" presupposes a lot of sci-fi hocus pocus about machines being instantly better than us at everything we do well, including the reduction of their own power consumption needs. It's crap and requires such a heavy suspension of disbelief on the part of the reader that it's on about the same level as cypherpunk fiction in the 1970s or giant robot cartoons. With the probably exception that I LIKE cypherpunk and bigass robots...Singularity is just paranoid screwiness.
Oh get off the high horse. Just because something is deprecated does not make it wrong. HTML v3.2 is going to last us a long, long time because it is a) ubiquitous b) easy to understand c) easy to implement. It is becoming the HTML people "learn in school," the vulgate language.
And you know what? It doesn't hurt anything. The page downloads fast enough, and it renders nicely. Slashdot is an informative, easy to use web site, not a pretty one. And that's the way we likes it.
(oh, and i think the style strip is due to some ancient browser vulnerability that was used to make goatse.cx pop up. you forget that just because you don't use the cool new features of CSS to do ungodly ugly things doesn't mean others won't. and CSS is still so damned complicated and varied by implementation that it's probably more beneficial to stip it out than it is to learn how to clean it.)
Wow. I wish I would have thought of that before I diss'd them on slashdot...but that would require a modicum of faith in the cultish and often divisive open source community.
Oh, and i modded myself down on this post because i'm a troll.
Besides which, you can't compare a VIA C2 to a G4. Hell, even comparing it to a G3 is sort of like comparing a scooter to a Harley. Of COURSE it's half the price...it's less than half the machine!
And it begs the question: why are we supposed to trust a company that can't even get it's own marketing straight? They're just a company with a name that's a bad pun on the most effectively marketed operating system ever...I'd put more stock in AmigaOS. At least Amiga is trying it's own thing.
It's not America, but human nature that's "consumerist." Haven't you heard some of the horror stories of child prostitution in Asia and Africa? The latter has a rampant problem with grown men raping young girls because of a common myth that sex with a virgin cures AIDS.
But of course, that's only what I read in the news, and I can't imagine African news is any more interested in truth and level headedness than our American FOX/CNN/MSNBC cloud
Romeo was 20 something and they got married first. Not so kinky, Jerry Lee Lewis did the same thing.
However, the Troma films version is decidedly kinky. Juliet's father dresses her in vinyl underpants and keeps her in a clear plastic cage.
The whole reason child porn is illegal is the common idea that children are easily misled. I'd like to say, as I did in high school, that I disagree and think that kids, teenagers especially, are capable of making their own decisions. But now I disagree.
Here't the thing: just having a naked picture of a child isn't going to incite anybody to do anything against their nature. It isn't the catalyst to the downfall of society and isn't going to create some massive underground rape culture.
But if it were legal -- as in legal to sell -- it would create an "industry." And because of the social stigma against child porn, it's not going to be a friendly industry. It's going to be made up of the worst dregs of the porn industry, full of trickery and lots of children would fall into it. Promised low incomes that sound fantastic to young ears they'll be easily pressured into doing things they don't want to do by people who don't care for them.
Making it legal to posess child pornography is like creating demand for a grey market in traffic and honestly making it worse for kids.
Do I think that pictures along are going to hurt society? No, and therefore I think sting operations and such are worthless as anything more than a deterrant. But come on. The world is dangerous enough with kids preying on each other sexually. We don't need to throw clever pervs into the mix as well.
Goddamn it, a picture of a baby in a bathtub IS child porn! I don't know where you parents get the idea that just because you find tiny naked kid parts cute that makes it okay to take pictures of them. Isn't it wrong to take naked pictures of a naked teenager? Cut the double standard. Naked baby pictures are sickening. I know I'm not the only person who things so.
Anne Geddes should go to prison for objectifying naked children. She's worse than Maplethorpe and Saudek combined.
And of course, if there is a list of sites to block, discerning porn fans from other states will get ahold of it and cut down their search times. What to do with the extra time? Why, trade on IRC. Censorship never works on the internet, it's way too wily
I read the British Columbia stopped it's pedophilia stings and pumped the money into an awareness program. I wonder how they're doing with that.
Case in point: Tom Pabst. The guy was the BEST reviewer of all forms of hardware in 1999. I wouldn't trust anybody else. The guy had the know how and the connections.
Then, he gave a couple bad reviews. To people like Intel. Suddenly, he wasn't getting the latest chips. Suddenly, he was borrowing them from other people, violating NDAs, or waiting for the release like the rest of us. His site became "great, but old news" instead of "just in time answers." I moved my eyes to the more "read between the lines" style of Anandtech.
Moral of the story? If you're a corporate whore, you'll benefit from it pretty much every time.
By the way, I love NT. Windows 2000 is the best system I've ever used and it's mostly because it developed organically. If linux had a bunch of eggheads worrying about ease of administration and intuitive GUIs instead of hooking everything through the command line and scoffing at developers who want to get paid, I wouldn't be tapping this into Internet Explorer.
Open systems have their own pitfalls though. They are far, far worse when it comes to bare bones implementations of hardware drivers because there really is no impetus to build a system that isn't immediately useful. Say what you will about windows, it's release quality. The parts of the API that don't work are on the fringes, new development, the same sort of thing that open system developers would change, too, if they didn't work the right way. Case in point: Apache 2, Cocoon 2, the infinite iterations of Freenet which are incompatible with previous nodes.
In the case of an open system, if I need to send data out of a widget, I build a driver to send data out. I don't spend any time at all on anything else. Another person who wants to suck data into a widget will have to edit the driver. There is no call for them to maintain any sort of version compatibility with mine; frankly, if they feel it's beneficial for their use to rape the core, they'll do it without any qualms. On the other hand, manufacturers of closed systems have to just that. They have to anticipate any way a person would use their driver, or else the hardware won't sell. Just ask Iomega about their Buz box, a device which never sold because they never released a 32 bit driver for it.
And in the end, since only one party's been touching the driver from a closed system, it's much cleaner and much more likely to have few points of failure. Fixing it is much faster because it's somebody's job to maintain that code. You don't have to download a source package touched by a hundred hands for a hundred reasosn and try to piece together what they each were trying to do. Of course, you have to rely on the vendor, but if you're an embedded systems developer you better damn well have a good relationship with your vendor as it is.
So the decision is thus: On one hand, you have open software. Where the freedom of use and lack of liability mean the possibility of a hundred apis broken off the same core with minor differences in the feature set which are all totally incompatible (PocketLinux). Where upgrading your webserver can break every other application on your machine (Apache) due to subtle dependencies for features you probably won't use and don't have the time to recompile for. Where continued improvement and a staunch refusal to freeze development so you can use it as an "API" forces you to maintain a constant watch on all new versions of all dependent code (GNOME, GTK, even glib). A platform where all of this work is placed on me, the developer, since it's too complicated for an IT person in our target market.
On the other hand, you have "POS" windows. A platform for which I can write a simple client server application in a day and have it run on millions of PCs without recompilation, and where most bugs are due to using the API in a way that isn't intended or necessary.
OSS rocks, but at its worst it's a maddening cob job. I like having the source when it's good, but there's a LOT of confusing shit as well. Better to put the responsibility for fixing a driver or api call on somebody who is both libel to me and who has experience in doing so.
The "Black Box" design theory abounds because of the freedom it offers programmers from the dark ages of having to know the underlying hardware intimately before anything could be accomplished. It's what allows programmers to devote all of their time to doing what matters rather than pouring over volumes of errata and arcana.
The reason Windows became so popular, for example, is because its API offered programmers a way to manipulate graphics without having to make graphics calls. Variation from driver to driver was of no concern, and shouldn't be -- that's an IT issue which can be repaired without redoing the entire application.
And in a perfect world, there's no problem. If a driver hooks into an API properly and documents any disperity, then the black box theory holds true. Problem is, driver aren't perfect. A lot of them are designed for bare bones functionality, and only optimized as necessary (hence how Nvidia's still squeezing substantial horsepower out of my ancient GeForce GTS with every new driver release). Obscure hardware cases always cause trouble, which is why Dells are (sometimes) more reliable than "no name" machines with "better" hardware. Dell has the clout to make sure the drivers are as seamless as possible.
What's the solution for embedded developers? Design and test the drivers in house, so the black box coders have a shoulder to cry on when hardware doesn't act properly. But it should not be the core developer's job to know what goes on with the hardware. That kind of thinking bloats budgets, increases the complexity of the project and ultimately the cost. Modularity, even though it makes things more difficult to map in total, makes things easier to deal with on a micro level. If the application works when unit tested but fails on the release machine, then it's the driver's fault. Much easier to fix than it is to perfectly replicate the release in your tests.
Expecting EVERY software developer to be an electrical engineer as well is absurd unless you intend to pay them for both degrees. Better to keep it modular and put the pressure on the hardware abstractor to do a good job of catching the tiger's tail.
Of course GOOGLE isn't buying it. Google's delivering nothing but text and small images. The technical bottleneck in everything Google does is bus speed and the internet itself. Logical solution? Parellelize the heck out of the problem and bollocks the "high tech."
.01 GHz faster than the last.
Guy I know. His father's company just bought a packed X-serve RAID. They don't keep a lot of records and they can't go paperless due to restrictions in their industry. So they essentially have purchased a 2.52 TB MP3 server. It'll almost fit all his Zappa bootlegs.
At the same time, there are LOTS of customers out there who need 64 bit. Lots of folks who need faster and more robust searching. Customers who are dealing with sluggish behemoths from IBM that the Itanium (or, I prefer, the Hammer) is perfect for. Aerospace engineers who need to perform simulations and stress tests. And regular cats who just want incredibly high resolution math for doing speech to text, photoshop or gaming.
But GOOGLE? Google could operate on pocket calculators if you got enough of them, yahoo too. The web is wonderfully low-tech and low-resource if properly designed.
I wish more companies would follow their lead and use whatever technology is best for their model rather than something big, flashy and ultimately too expensive. This would lead to higher margins. What? Yeah, if you have fewer people clamoring for the TOL, you can work slower and charge more. Think Apple. Think Porsche. Then there's less of a need to farm everything overseas and release a new clock speed every month that's
I've been told that my credit is bad because of the number of times i've been denied refinancing loans on my car.
I try for these OFTEN because i'm two years into ownership on a car at 13%, pretty low at the time. I have excellent credit besides this (own a house, missed one payment in 7 years, high income to debt ratio).
Refinancing a car isn't like refinancing a home...a lot of banks don't want you to refi if a car is too old, because there's so much more risk involved. After all, if I defaulted and they repo'd the thing i'd still owe something on the order of 4k they'd never see. That 13% then is fucking me...unless i overpay, which i'm not willing to do, the point at which the car is worth more than the outstanding payoff remains in the distance, somewhere around summer of 2004. So there's all this risk, even with my bad credit, and no real reason for the bank to accept considering how little they'll make off the deal. And each rejection leads to another reason to reject me.
I'm not great with money but i'm also not a derelict. If only equifax would understand this!