I have always considered motherboards THE most complex, interesting, and difficult part of a computer to design, and design well.
Every single component on a motherboard can be a magnificient piece of work, but it deals only with a limited number of variables. A CPU or chipset component ultimately only deals with bus protocols and internal design. The same goes for other components like memory or harddrives.
But everything comes together through the motherboard, and that (in my eyes) makes it the hardest part of a system to design well, considering the number of variables. A truly well done MB design is really a piece of art.
Let me guess: users generally prefer stuff that boasts a lot of features, over stuff with fewer features that work reliable?
I think there's a psychological explanation here (I kid you not): researchers found some time ago, that people attach more weight to opportunities missed, than to opportunities grabbed. If you had a chance to earn $100, and skipped it, you feel bad about missing that chance, more than you feel good if you grabbed the chance and earn the money. It's like bad publicity sticking longer to a brand name than good news involving that brand.
Features are sexy, reliability isn't, even if it's important. I hope that for software at least, times are changing. But don't count on it.
For a company like Sun, profit would be: profit from (hardware sales + software licenses + services/support) * numbers. High price = low volume, low price = high volume. Maximum profit is an optimal point somewhere in between.
Open sourcing Solaris could improve it in general (and at a lower cost to Sun!). And making (hardware + software solution) more attractive to customers, could drive up sale volumes. So my guess is that this is just Sun's attempt to move closer to that optimal price/volume point.
From the article: "WAV has one of the lowest compression ratios and is virtually lossless, but it is not streamable"
WAV isn't compressed format at all (check filesize against audio data rate), IS fully (not virtually) lossless, and although the format wasn't designed for it, being raw audio data, ofcourse you could stream it.
I think current DVD players are a good example of the practical 'use' of DRM. They contain protection mechanisms (in hardware and software) like region coding, CSS and more. What does it do for consumers?
There was a Slashdot story earlier about an interview with MPAA's Jack Valenti, who said: "I really do believe we can stuff enough algorithms in a movie that only the dedicated hackers can spend the time and effort to try to plumb through those 1,000 algorithms to try to find a way to beat it". He really doesn't have a clue, does he? A consumer gets the content, has a decoder, and gets decoding keys where needed (somehow), and to prevent interception, decoding will have to be done at the endpoint, the consumers' equipment. And then you expect to be able to ensure that content never leaves the device once decoded? Get real.
Consumers are faced with extra hassle, limitations resulting from DRM features, and building those features into equipment and software makes everything more complicated and expensive. Hackers on the other hand (both black hat & white hat), will have circumvented DRM features in no-time.
The interesting point here: the barrier it presents to hackers is removed quickly, and isn't an issue after that, but all disadvantages it presents to consumers, REMAIN. If, in 10 years from now, you want to write a software DVD player, chances are at some point you'll still have to deal with region codes, CSS and other useless crap, and DVD-enabled equipment will always be more complicated and expensive than it has to be because of the included DRM features.
Concluding: DRM just adds useless overhead, extra cost, and doesn't do squat to prevent unintentional copying (aside from whether you think it should). I wrote a rant titled "CONTROL versus FREEDOM" some time ago, that isn't of much interest anymore, but its conclusion still holds. For me, it means that I won't invest any money in products that have significant DRM features built in. CD's with copy protection? Game consoles that you're not allowed to mod, or run your own software on? Stick it up your.....
Some may think a good approximation of a calculation problem is "good enough". Too many variables or numbers? Just throw a more powerful calculator at the problem.
Not so. Exact solutions like the ones provided by mathematical formulas are still useful for a number of reasons:
Using an exact formula can provide a shortcut, that enormously reduces the amount of calculation you have to do. It also allows to do some calculations by hand.
More important: an exact solution provides insight in the relation between its variables. That's very important from an educational point of view. And by substituting in other formulas, this can also advance the state of the art in other areas of science. A numerical solution, or a progression towards certain values might help you believe there is a certain relation between variables. An exact solution can help you prove such a relation. That is very significant for any sort of theoretical science.
(for those who can't read articles in Dutch)
Well, more exactly: in math-speak, which I understand somewhat, but not quite enough to fully understand the details of the method.
Anyway, I understand the problem, and that this is a general solution to it, and advances an age-old mathematical problem where no significant progress was made in over a century. So: cool! And kudoz to its inventor!
Summarised: taxpayers' money is used to put satellites in orbit (or obtain imagery from 3rd parties), so shouldn't that same taxpaying public have full access to images that are sent back?
I could understand some exceptions for super high resolution pix of military installations or classified research facilities, but all images? That is way too broad, and no justification for it. Besides, if you have something secret that can't bear daylight, then just hide it/cover it up, like normal people do.
Gerard Beekmans is the guy who started the Linux From Scratch project. It's not one of the most popular distro's, but I'm pretty sure it's an important project in terms of inspiration, useful info, and generally helping Linux conquer the world.
From the article: "Lexmark shares were down $3.24, or 3.8 percent, to $82.86, in midday trading on the New York Stock Exchange"
I find it weird that people (or stockholders) would react so negatively to news like this. Making millions of consumer electronic items, there will always be some mistakes made. What matters is not that shit happens, but how it is handled.
Ofcourse it would be nice if products were perfect, but I don't regard a manufacturer any worse because of a mishap like this. If they handle such a recall in a painless and professional way, that would more likely boost my confidence in them. Problems like this are often very minor things, and say nothing about the overall quality of their product.
I would think worse (and less likely buy something) of a company that tries to look good, sweep problems under the rug, and pretend nothing happened.
Microsoft might as well let SenderID go open source.
I really like development of this whole MS initiative so far, being shot down by Apache, Debian, and now reconsidered for open sourcing?
It is pressure being applied to MS. And it's not pressure from Linux zealots, or from FSF supporters, but because it's necessary, and useful for this SenderID to work. In other words: free market pressure.
End users get what they need (nothing at all, or something open, that might even work), and this possible solution to the spam problem is judged on practical and technical merits, rather than a big organisation pushing it.
Yes, still running Win98 here, and I have the same experience. Visited Windows Update after install, then stripped out IE (98lite), full backup, use Mozilla, regularly updated virusscanner, and rarely run binaries fresh of the 'net. Result: last worm infection was long ago (on a LAN party), lockups are rare, no weird problems of any kind.
I guess a major factor is that many exploits are created by reverse engineering patches. As Microsoft has ended active support for Win9x systems, that also means no new patches for hackers to reverse engineer. Then there was this source code leak, wasn't it Win2k source code? So different code from what's in Win9x. And as Win9x systems are replaced with Win2k/XP, their smaller market share makes Win9x a less interesting target.
I am curious why other systems manufacturers like Dell, Compaq, or Hewlett Packard, do not overclock their products
Because many of those maufacturers thrive by the "low prices, high volumes" concept.
Overclocking enough to notice, and without losing to much stability, requires quality components, careful tweaking of hardware, BIOS and software configuration. The first thing a normal user would do, is install some stupid proggie that interferes with that careful tweaked system, and cause helpdesk headaches for the company that sold them the box. Simply put, overclocking has the opposite effect from what's needed for this low prices, high volume market. And warranty issues make things worse. That's why overclockers are generally a different crowd than average home users.
How much time before you can buy something similar for half the money? 3 months? 4? I can understand you're willing to pay more for the fastest available, but eehh.. to do what? Gaming? Get a quality motherboard, fast memory and one of the latest videocards. Much cheaper. Scientific calculations? Get a small blade server or mini-cluster, that'll better suit the job. Just to impress your friends? Get a nice looking case, do some modding, add some cool looking lights. Much more fun.
Managers in many companies are expecting emails to be returned whenever. 7am 9am 2pm 7pm 10pm
Yes, this is normal for many companies, and it's stupid. E-mail never was, or is, meant to be answered immediately, if at all.
The whole concept of e-mail is that the receiver can read it, and decide how to act upon it, when he/she wants, like with snail mail. If you want to communicate with someone but don't have patience to wait for a response, then don't use e-mail, period. If you want immediate response, go get someone on the phone, or go to their office and talk to them in person. Use the right tool for the job.
I usually read e-mail several times a day, but for this same reason I don't mind NOT reading e-mail for a week (or 2), when I'm on vacation, or just don't feel like browsing through the latest 419 scams. The thing I like about e-mail is that it gives you, and the other party, the time to come up with a well-constructed response. For instance in tech support, the customer can provide all relevant details in one go, and the company can take some time to talk to manufacturers or head offices, during normal office hours.
Hey what happened? I just read the article ON THE NY TIMES site, yet I never did register there. Did the NY Times change their policy from "Get our logo tattood on your behind req'd" to "Let's make an exception for all those nice/. readers", or what?
My initial thought too: what the hell is it? But reading about it, I realise I've come across some local initiatives like it, in the past. Rough description:
In a normal job, you get paid normal currency for an agreed upon amount of work (like a number of $$/hour).
With voluntary work, you do work, but don't get money in exchange. Rewards will be non-monetary, say having a satisfied feeling of having improved the world a little bit. For some programmers, writing Free software could be in this category.
I would describe Timedollars as something in between. You do work, don't get paid real currency, but you do get rewarded in a defined manner. At a later time, that reward can be paid back by work that someone else does for you.
It's similar to when you help someone move furniture, and he/she pays you back later by inviting you to a barbeque. Two important differences here:
A 'unit' in which the amount of your work is measured, is clearly defined.
It's not an exchange between 2 individuals, but work and rewards are exchanged between larger groups of people. For that to work, you need some organisation to help negotiate 'jobs', and bookkeeping of earned credits.
If you like an artist, I mean REALLY like an artist, you will be happy to pay for their music
I'm sorry to say, but you wasted a large part of that 75 dollars of yours. A big chunk of it was taken by distributors, record company, songwriter, and possibly others. Only a (possibly small) part ends up on the artist's bankaccount.
You may consider asking an artist if it's somehow possible to donate to them directly. If they say: "sorry we can't take your money" (stupid!...), go buy their CD. But if they say "Sure! Here's my bankaccount number" (smart!... they've seen the light), you determine the easiest method of getting their music, you decide what you feel their work is worth, and the artist gets 100% of that.
Consider management style jobs: lots of things to do, many interruptions, and s**tloads of information coming from everywhere at the same time.
The trick is not to let your life be run by the outside world, but interface with the outside world in a way that works best for you.
You get too many people calling you on the phone, or e-mailing you more than you have time to spend on answering them all? Get a secretary, and let him/her filter phone calls, e-mail or paperwork.
You want to focus on a particular activity? Tell people you work with, and don't ALLOW any interruptions during that activity. Business meetings are an example. Another example: I don't like being on the phone while I'm in busy traffic, or in the middle of shopping in a supermarket. Solution: I just leave my phone at home when I go out to buy groceries.
Don't let people grab your time whenever THEY feel like it, but give people the opportunity to do so when YOU're ready for them. In CPU terms: it might work better for you not to work in an interrupt-driven style, but use a polling method instead.
Actually, the current state of IP laws could even act as catalyst to increase adaption of FOSS by the masses, and (better yet) the principles they're based on. How? When consumers actually feel the effects of IP laws as negative.
IP law is accepted by the public, because a) in many cases, they aren't/can't be fully enforced (example: warez/MP3), or b) the net effect is perceived as small, non-significant (like in the media player/music/movie content business).
Free/Open Source remains largely 'under the radar' because consumers still don't feel they are hurt much by non-free infrastructure (like closed-source operating systems or patent-encumbered file formats/protocols), or Digital Rights Management (like DVD region coding/CSS). For the average consumer, this provides little incentive to adopt products that support open technology.
But when IP law is more aggresively pursued & enforced, alternatives automatically become more visible, and attractive. DRM may be accepted as long as it stays a minor inconvenience, a small annoyance. But as soon as it actually prevents people from doing things, it makes non-DRM encumbered products more interesting.
I'm not too worried if M$ decides to put more DRM features into Longhorn. That just makes Open Source software a more attractive alternative to consumers. Like I'm not bothered much by high oil prices. That will just make fuel-efficient cars and environment-friendly energy sources more interesting. A small economic crisis can help there too.
So in the long run, IP laws will largely act to shoot themselves in the foot. It's just sad how much hassle they cause in the meanwhile.
Slashdot Mirror
Every single component on a motherboard can be a magnificient piece of work, but it deals only with a limited number of variables. A CPU or chipset component ultimately only deals with bus protocols and internal design. The same goes for other components like memory or harddrives.
But everything comes together through the motherboard, and that (in my eyes) makes it the hardest part of a system to design well, considering the number of variables. A truly well done MB design is really a piece of art.
If you're stressed out, and a beer helps you relax at the end of a day, that stress-removal is a positive health effect in itself.
I think there's a psychological explanation here (I kid you not): researchers found some time ago, that people attach more weight to opportunities missed, than to opportunities grabbed. If you had a chance to earn $100, and skipped it, you feel bad about missing that chance, more than you feel good if you grabbed the chance and earn the money. It's like bad publicity sticking longer to a brand name than good news involving that brand.
Features are sexy, reliability isn't, even if it's important. I hope that for software at least, times are changing. But don't count on it.
Open sourcing Solaris could improve it in general (and at a lower cost to Sun!). And making (hardware + software solution) more attractive to customers, could drive up sale volumes. So my guess is that this is just Sun's attempt to move closer to that optimal price/volume point.
WAV isn't compressed format at all (check filesize against audio data rate), IS fully (not virtually) lossless, and although the format wasn't designed for it, being raw audio data, ofcourse you could stream it.
There was a Slashdot story earlier about an interview with MPAA's Jack Valenti, who said: "I really do believe we can stuff enough algorithms in a movie that only the dedicated hackers can spend the time and effort to try to plumb through those 1,000 algorithms to try to find a way to beat it". He really doesn't have a clue, does he? A consumer gets the content, has a decoder, and gets decoding keys where needed (somehow), and to prevent interception, decoding will have to be done at the endpoint, the consumers' equipment. And then you expect to be able to ensure that content never leaves the device once decoded? Get real.
Consumers are faced with extra hassle, limitations resulting from DRM features, and building those features into equipment and software makes everything more complicated and expensive. Hackers on the other hand (both black hat & white hat), will have circumvented DRM features in no-time.
The interesting point here: the barrier it presents to hackers is removed quickly, and isn't an issue after that, but all disadvantages it presents to consumers, REMAIN. If, in 10 years from now, you want to write a software DVD player, chances are at some point you'll still have to deal with region codes, CSS and other useless crap, and DVD-enabled equipment will always be more complicated and expensive than it has to be because of the included DRM features.
Concluding: DRM just adds useless overhead, extra cost, and doesn't do squat to prevent unintentional copying (aside from whether you think it should). I wrote a rant titled "CONTROL versus FREEDOM" some time ago, that isn't of much interest anymore, but its conclusion still holds. For me, it means that I won't invest any money in products that have significant DRM features built in. CD's with copy protection? Game consoles that you're not allowed to mod, or run your own software on? Stick it up your .....
A mod point gives you the freedom to mod this up as "Funny".
Not so. Exact solutions like the ones provided by mathematical formulas are still useful for a number of reasons:
Well, more exactly: in math-speak, which I understand somewhat, but not quite enough to fully understand the details of the method.
Anyway, I understand the problem, and that this is a general solution to it, and advances an age-old mathematical problem where no significant progress was made in over a century. So: cool! And kudoz to its inventor!
Summarised: taxpayers' money is used to put satellites in orbit (or obtain imagery from 3rd parties), so shouldn't that same taxpaying public have full access to images that are sent back?
I could understand some exceptions for super high resolution pix of military installations or classified research facilities, but all images? That is way too broad, and no justification for it. Besides, if you have something secret that can't bear daylight, then just hide it/cover it up, like normal people do.
Gerard Beekmans is the guy who started the Linux From Scratch project. It's not one of the most popular distro's, but I'm pretty sure it's an important project in terms of inspiration, useful info, and generally helping Linux conquer the world.
I find it weird that people (or stockholders) would react so negatively to news like this. Making millions of consumer electronic items, there will always be some mistakes made. What matters is not that shit happens, but how it is handled.
Ofcourse it would be nice if products were perfect, but I don't regard a manufacturer any worse because of a mishap like this. If they handle such a recall in a painless and professional way, that would more likely boost my confidence in them. Problems like this are often very minor things, and say nothing about the overall quality of their product.
I would think worse (and less likely buy something) of a company that tries to look good, sweep problems under the rug, and pretend nothing happened.
I really like development of this whole MS initiative so far, being shot down by Apache, Debian, and now reconsidered for open sourcing?
It is pressure being applied to MS. And it's not pressure from Linux zealots, or from FSF supporters, but because it's necessary, and useful for this SenderID to work. In other words: free market pressure.
End users get what they need (nothing at all, or something open, that might even work), and this possible solution to the spam problem is judged on practical and technical merits, rather than a big organisation pushing it.
I guess a major factor is that many exploits are created by reverse engineering patches. As Microsoft has ended active support for Win9x systems, that also means no new patches for hackers to reverse engineer. Then there was this source code leak, wasn't it Win2k source code? So different code from what's in Win9x. And as Win9x systems are replaced with Win2k/XP, their smaller market share makes Win9x a less interesting target.
My internet connection uses their network, so it may become sluggish for some time, while the /.ing is in effect...
Because many of those maufacturers thrive by the "low prices, high volumes" concept.
Overclocking enough to notice, and without losing to much stability, requires quality components, careful tweaking of hardware, BIOS and software configuration. The first thing a normal user would do, is install some stupid proggie that interferes with that careful tweaked system, and cause helpdesk headaches for the company that sold them the box. Simply put, overclocking has the opposite effect from what's needed for this low prices, high volume market. And warranty issues make things worse. That's why overclockers are generally a different crowd than average home users.
How much time before you can buy something similar for half the money? 3 months? 4? I can understand you're willing to pay more for the fastest available, but eehh.. to do what? Gaming? Get a quality motherboard, fast memory and one of the latest videocards. Much cheaper. Scientific calculations? Get a small blade server or mini-cluster, that'll better suit the job. Just to impress your friends? Get a nice looking case, do some modding, add some cool looking lights. Much more fun.
Yes, this is normal for many companies, and it's stupid. E-mail never was, or is, meant to be answered immediately, if at all.
The whole concept of e-mail is that the receiver can read it, and decide how to act upon it, when he/she wants, like with snail mail. If you want to communicate with someone but don't have patience to wait for a response, then don't use e-mail, period. If you want immediate response, go get someone on the phone, or go to their office and talk to them in person. Use the right tool for the job.
I usually read e-mail several times a day, but for this same reason I don't mind NOT reading e-mail for a week (or 2), when I'm on vacation, or just don't feel like browsing through the latest 419 scams. The thing I like about e-mail is that it gives you, and the other party, the time to come up with a well-constructed response. For instance in tech support, the customer can provide all relevant details in one go, and the company can take some time to talk to manufacturers or head offices, during normal office hours.
Hey what happened? I just read the article ON THE NY TIMES site, yet I never did register there. Did the NY Times change their policy from "Get our logo tattood on your behind req'd" to "Let's make an exception for all those nice /. readers", or what?
In a normal job, you get paid normal currency for an agreed upon amount of work (like a number of $$/hour).
With voluntary work, you do work, but don't get money in exchange. Rewards will be non-monetary, say having a satisfied feeling of having improved the world a little bit. For some programmers, writing Free software could be in this category.
I would describe Timedollars as something in between. You do work, don't get paid real currency, but you do get rewarded in a defined manner. At a later time, that reward can be paid back by work that someone else does for you.
It's similar to when you help someone move furniture, and he/she pays you back later by inviting you to a barbeque. Two important differences here:
I'm sorry to say, but you wasted a large part of that 75 dollars of yours. A big chunk of it was taken by distributors, record company, songwriter, and possibly others. Only a (possibly small) part ends up on the artist's bankaccount.
You may consider asking an artist if it's somehow possible to donate to them directly. If they say: "sorry we can't take your money" (stupid!...), go buy their CD. But if they say "Sure! Here's my bankaccount number" (smart!... they've seen the light), you determine the easiest method of getting their music, you decide what you feel their work is worth, and the artist gets 100% of that.
The trick is not to let your life be run by the outside world, but interface with the outside world in a way that works best for you.
You get too many people calling you on the phone, or e-mailing you more than you have time to spend on answering them all? Get a secretary, and let him/her filter phone calls, e-mail or paperwork.
You want to focus on a particular activity? Tell people you work with, and don't ALLOW any interruptions during that activity. Business meetings are an example. Another example: I don't like being on the phone while I'm in busy traffic, or in the middle of shopping in a supermarket. Solution: I just leave my phone at home when I go out to buy groceries.
Don't let people grab your time whenever THEY feel like it, but give people the opportunity to do so when YOU're ready for them. In CPU terms: it might work better for you not to work in an interrupt-driven style, but use a polling method instead.
Don't worry. Locusts eat plants, not data.
IP law is accepted by the public, because a) in many cases, they aren't/can't be fully enforced (example: warez/MP3), or b) the net effect is perceived as small, non-significant (like in the media player/music/movie content business).
Free/Open Source remains largely 'under the radar' because consumers still don't feel they are hurt much by non-free infrastructure (like closed-source operating systems or patent-encumbered file formats/protocols), or Digital Rights Management (like DVD region coding/CSS). For the average consumer, this provides little incentive to adopt products that support open technology.
But when IP law is more aggresively pursued & enforced, alternatives automatically become more visible, and attractive. DRM may be accepted as long as it stays a minor inconvenience, a small annoyance. But as soon as it actually prevents people from doing things, it makes non-DRM encumbered products more interesting.
I'm not too worried if M$ decides to put more DRM features into Longhorn. That just makes Open Source software a more attractive alternative to consumers. Like I'm not bothered much by high oil prices. That will just make fuel-efficient cars and environment-friendly energy sources more interesting. A small economic crisis can help there too.
So in the long run, IP laws will largely act to shoot themselves in the foot. It's just sad how much hassle they cause in the meanwhile.
It includes Linux as underlying OS to run on, and several components that are also used in Linux-based systems. Is that good enough?