Michael A. Polizzi, an assistant superintendent, said the district carefully researched future demand for jobs, examined college programs and surveyed students about their interests before settling on its first six majors: sports management, fine and performing arts, health sciences, international studies and global commerce, communications and new media and or liberal arts. I think I would have difficulty choosing a subject to specialise in at age 14, not just because I would be 14, but also because those options frankly sound like bullshit.
On the other hand, the school, Dwight Morrow High School, shares its campus with "Academies@Englewood":
"With more than 400 high-achieving students enrolled, the program --housed in its own building on the Dwight Morrow campus -- offers concentrations in engineering, law and public safety, biomedicine, finance, and information systems." I have to say, I would rather be in the high-achievers school with the decent subjects than the low-achievers school on the same campus which has such a poor selection of subjects.
Having had my apartment robbed, I can tell you that I did call the police. They sent a guy out over 24 hours later, who basically said "yeah, lots of people have been getting robbed around here lately", (note: i lived in a pretty nice area), "you probably won't get anything back. I hope you're insured." and left. No finger printing, no looking at the busted door, no follow up.
I, too, have had several similar interactions with the police - house break ins, car break ins, cars being smashed up, and suchlike.
What you've got to remember is they aren't "the department of useless fingerprinting to make you feel better". Any serious burglar would be wearing gloves. You say you live in a nice area, so it wasn't opportunistic crime, so it's almost certain they had gloves.
And if you think about it it makes sense: A burglar who leaves incriminating evidence at a crime scene and gets jailed for 6 months can only commit two crimes a year. A burglar who leaves no incriminating evidence at a crime scene could easily commit 200 crimes a year. So if you've been burgled, it's 100 times more likely to be the latter burglar than the former.
As I say, I've been in a similar situation; in one instance a guy decided to walk up the street smashing the windows of all the cars he passed. It's frustrating, and He Shouldn't Be Allowed To Get Away With That, but there was nothing to even indicate a suspect, let alone prove their guilt in court. What are the police supposed to do?
There are some things that could be done about this: We could put good quality CCTV cameras watching and recording everything so criminals can be found later; we could fit GPS trackers to all citizens so past movements are logged and can be searched to find suspects; we could put 'a policeman on every corner' literally, so that crime will be seen; we could get rid of probably cause for searches so after burglaries the police can just search the homes of people who have previous convictions, looking for the stolen goods; we could segregate people with a financial incentive to steal into slums which they are not permitted to leave. However, society (justifiably) considers these options unacceptable.
Anyway, here's my point: All the police can achieve once a crime is over is to give you a crime number for your insurer. It's frustrating, but it's reality. Sorry.
I have no idea how complex an OpenSPARC is, but I assume it is something equivalent to an ARM9 or so and will fit in a $10-or-so FPGA. Well, Sun's press release says:
the UltraSPARC T2 processor delivers an unprecedented level of integrated system functions on a single chip [...] Eight cores and eight threads per core [...] Dual, virtualizable, multithreaded 10 Gigabit-per-second Ethernet ports [...] Eight floating point units [...] Eight lanes of industry-standard PCI Express I/O [...] Quad memory controllers deliver more than 50 Gigabytes-per-second of memory access [...] The UltraSPARC T2 processor is available in production quantities this quarter, with prices starting well below $1,000, and licensing options wide open for derivative works. [...] Sun UltraSPARC T2 @1.4GHz (64 threads, 8 cores, 1 chip) For comparison, ARM's FPGA-targeted Cortex-M1 is a single-threaded single core processor, and on a nice Xilinx FPGA you can get 170MHz out of it; in Actel's FPGAs (where you can get the core free-as-in-beer) you don't get that clock speed. ARM's core also lacks hardware floating point support - in fact you don't even get hardware integer divide support. However, Cortex-M1 is very small, making it cost-effective.
On the other hand Cortex-M1 is based on a cut-down version of a silicon core - so there might be scope to cut down the OpenSPARC to a single-threaded single core for FPGA. But at this point I'd be surprised if anyone was running OpenSPARC anywhere except in simulation.
Just my $0.02
p.s. How are you involved with ARM cores on FPGAs at the moment, if you don't mind me asking?
If you fill your shop with 15 average Java developers, paying an average of $60k per developer you have an approximate labor cost of $900k/year for your development staff. Not considering any non-salary benefits.
Suppose you instead took the time to find 5 expert, or at least above average, Perl developers at $120k each per year. That seems to be the gist of the article, and it's a pretty reasonable conclusion: Experts can be very much more productive than non-experts.
However, it is also my experience that it isn't always easy to tell highly capable people from the merely capable; that is, I've worked with people who seemed very good initially, but in the fullness of time I realised they were not. And that, of course, is a benefit for having 15 developers instead of five: Any given hiring mistake costs half as much, and reduces your workforce by a fifteenth instead of a fifth.
So how are you supposed to find these expert programmers, and how can you tell a $60k developer from a $120k developer? By asking brain teasers like Microsoft and Google are reputed to do?
Maybe, just maybe, the different genders gravitate to the fields that they like. Or, gasp, are suited for. I agree that may be the case.
However, several professions were at one time male-dominated, but have changed with time. One such profession is medicine; see the study Women in hospital medicine in the United Kingdom - in particular this graph. As you can see, in 1960 20-25% of medical school entrants were female; by 1990 50-55% of medical school entrants were female.
Similarly, consider the legal profession. See the (somewhat old) study Women in the legal profession : theory and research, particularly Table 2 (page 3). In 1961 3.9% of Australian law professionals were female; by 1991 this rose to 25% (full time only).
Anyway, here's my point: Some historically male dominated professions have changed to be less male dominated. The same might happen in IT. Furthermore, the inference that current gender distributions imply fundamental gender-dependent abilities may be incorrect.
Of course, IT doesn't enjoy the pay and status of the medical and legal professions, and so may not experience the changes seen in law and medicine.
In physics, the Casimir effect or Casimir-Polder force is a physical force exerted between separate objects, which is due to neither charge, gravity, nor the exchange of particles, but instead is due to resonance of all-pervasive energy fields in the intervening space between the objects. [...] Since the strength of the force falls off rapidly with distance it is only measurable when the distance between the objects is extremely small. On a submicron scale, this force becomes so strong that it becomes the dominant force between uncharged conductors. Indeed at separations of 10 nm -- about a hundred times the typical size of an atom -- the Casimir effect produces the equivalent of 1 atmosphere of pressure (101.3 kPa).
You would think scheduling a big upgrade for the middle of the holiday season would be asking for trouble.
What's wrong with saying "the second weekend in February" or some similarly random date? It's a weekend so it won't interfere with business, but unlike new years day it won't mess with employees' personal lives too much.
There's a reason businesses and governments don't start their financial/tax years on the first of January, after all.
It was always my understanding that for a 'bubble' to 'burst' there has to be a market full of overvalued assets. For example, people valuing pet food delivery companies at millions of dollars. I'm not sure what particular overvalued assets this bubble is made of.
So what does Dvorak say?
The current bubble, already called Bubble 2.0 to mock the Web 2.0 moniker, is harder to pin down insofar as a primary destructive theme is concerned. A number of unique initiatives, however, are in play here. Let's look at a few of the top ideas floating the new bubble.
Neo-social networking. Today everything from YouTube to the local church has a social-networking angle. And this doesn't even consider the actual social-networking sites, from MySpace to LinkedIn to Facebook to even Second Life. This scene is totally out of control and will contribute to the collapse for sure. MySpace was purchased by fox for a somewhat excessive sum, but that's already happened. Facebook's owners reportedly want two billion dollars for the company, but no-one has paid them that. So from this category I see one company, facebook, and it isn't even publicly traded.
Video mania. With dozens and dozens of YouTube clones cropping up to get on the "throw money away" bandwagon, you must sense that the eventual shakeout in this space will have a negative impact. Youtube was a rather expensive purchase for Google and it's hard to see where the payback is, but that's already happened. I can't really think of any competitors anyone is likely to invest in... google video, perhaps, but that's owned by Google anyway.
User-generated content. This idea has been around since Usenet and just keeps improving. It will make no contribution to the overall collapse except for users reporting the collapse. "This part of the bubble is not part of the bubble"??
Mobile everything. Here is another concept that has been in play since the mid-1990s. It cannot trigger a collapse since it will never fully get off the ground, although the iPhone mania may be a bad sign of something. Mobile what? Are mobile phones a bubble? Or is there a bubble of iPhones and iPhone-like-devices that I'm not aware of?
Ad-leveraged search. Most search engines will fail as a matter of course. This segment of the industry is mundane. It would be affected by a crash but not trigger one. You mean Google?
Widgets and toolbars. I cannot see the widget scene going crazy, and the jury is still out on toolbars. But there is the potential for nuttiness, I think. The problem here is that these things tend to be dependent on the stability of operating systems and browsers. One bad operating-system patch and suddenly nothing works. There's a "widget and toolbar" bubble? I don't know of any company built around selling "widgets and toolbars".
So, here's Dvorak's bubble of over-valued assets:
MySpace = fox Facebook - privately owned Youtube = google Google iPhone = apple
Or in other words, the best examples Dvorak has of the bubble of the late 90s repeating itself are:
Fox Google Apple Facebook
Personally I don't agree with Dvorak's assessment that these companies are about to collapse (although it seems unlikely anyone would pay $2 billion for facebook).
many will 'either stick with the Windows they have, or turn to Linux or Mac OS X' Well, lets see what the actual numbers are (quoting the article):
2% said they are already running Vista 9% said they planned to roll out Vista in the next three months. 87%, said they would stay with their existing version(s) of Windows. 8% of those polled acknowledged Linux plans and 4% said they would deploy Mac OS X.
I would say "many will stick with the Windows they have", certainly, but I'm not sure I would call 8% or 4% 'many'. And somehow I suspect 'linux plans' might not mean complete replacement of Windows on the desktop.
4) What is the cost of the car and cost for the replacement battery pack? --> Not yet known for USA, in Norway 200.000 NOK 200,000 Norwegian kroner = 34,130.80 US$ (= 16,740.63 UK£)
Well, like it or not, a humanities degree is cheap compared to engineering or science. All that lab equipment (and space) costs money, not to mention the people who set it up and keep it running.
True, but surely an engineering department could more easily enter into partnerships with companies, make profit from patents/discoveries/spin-outs, and suchlike?
Humanities departments, you'd think, would mostly get their research funding from the government, because they don't really have a useful product to sell. An engineering school, on the other hand, is probably producing research with a high value.
This doesn't have to mean patents and secrecy; just that if someone discovers something useful, they can work with industry providing experience/expertise in exchange for money. The university/engineer gets money; the company gets a head-start on its competitors; national government collects more tax as companies become more competitive; and humanity as a whole benefits from the new technology getting out ASAP. Sounds like a win-win situation to me.
I'm not sure humanities people can do this much - the latest thoughts on film theory or marxism may be fascinating, but cannot be converted into money in this way.
It's my guess that, in the next 4-5 years, LVDS will be supplanted by DisplayPort in all the "big 5" LCD manufacturers (LG/Philips, Sony/Samsung, CMO, AUO, and Sharp). AMD/ATI, nVidia and Intel mobos/GPUs will likely adopt this on a bigger scale starting next year. The one thing that's for sure is that all of the manufacturers not aligned to Silicon Image (read: everyone) are hell-bent on pushing through DisplayPort, no matter how painful or how long it takes. I'm not so sure - at present I'm using DVI to connect a $100 graphics card to a $300 flat panel; one would think licensing costs would be a negligible fraction of the final selling price.
Furthermore, one would think a device supporting only DisplayPort would command a lower price than one supporting both DisplayPort and DVI, because many people have DVI hardware already, and that price premium would be greater than the licensing fee for DVI.
To me the idea of DisplayPort displacing DVI is similar to the situation with GIF and PNG - namely an established standard with a license fee vs. a less established standard with no license fees*. The outcome of GIF vs PNG was that software gained support for both standards, some people continued to use GIF for backwards compatibility and from habit, and the patents on GIF expired before users stopped wanting GIF support.
I can imagine a similar ending for DisplayPort - namely never displacing DVI enough to save manufacturers any money.
Could someone explain to me, preferably without recourse to religious argument, what is wrong with these kids viewing porn?
Nothing inherently, but in the particular case of OLPC I can see why filtering might be reasonable:
1. In a class room situation any web browsing could be disruptive to teaching, but pornography particularly so. It would be pretty weird if at work the guy opposite me in the office was looking at porn all the time!
2. Parents may be reluctant to give children access to OLPC machines if the machines have a reputation as 'porn portals'. Non-adoption would obviously prevent the anticipated educational benefits of OLPC being realised. To put it another way, revolutionising religious views of pornography is not part of OLPC's core aims.
3. Similarly, if looking for government funding in the US, it's probably useful not to have the stigma of pornography hanging over your head - what with all the religion involved in US politics etc.
If I look over your garden fence and see that you're building a giant widget and then you notice and offer me a tour of your giant widget do you have any legal recourse if I decide that I like the idea of having my own giant widget and then make one for myself?
It sounds like you might have just deduced why legal constructs like 'patents' exist in many countries:-)
Furthermore, if you write to television licensing and tell them your TV is used for gaming/DVDs only they will stop sending you letters; and if an inspector visits and you show them the TV connected to the games console, they will say "that's fine, you don't need a license". I have verified the two preceding facts by direct experience at two different houses within the last five years.
1. Park my bike, shower etc. 2. Turn on computer. 3. While it starts, get a coffee. 4. Log in, drink coffee, check e-mail/calendar. 5. Get to work.
I've got to say, that sounds like the sort of interview question that would get some pretty boring responses. Like mine, above. So I usually jazz it up a bit in interview:
1. Park my unicycle, change out of my superhero unitard. 2. Get a new guitar from the IT guys because I smashed mine at the end of my last performance. 3. Check in with each of the 10,000 people who work under my command, all of whom I know by name. 4. Have my executive assistant relay my e-mails to me, one character at a time, by throwing lettered frisbees back and forth between my company's two tower blocks. 5. Take my second breath of the day.
So far I haven't had any job offers, but I figure the market is pretty competitive at the moment - it's only a matter of time!
What's wrong with the integrated graphics of the Macbooks? I know a guy who plays WoW on his Core2 Macbook, and he gets better framerates than his (admittedly aging) PC desktop. Well, it's an Intel GMA 950. Here are some reviews: 1, 2.
As far as I can tell that first review says at 640x480 you get 6.4 frames per second in Half-Life 2. The second review lists the GMA 950 as not performing very impressively, though it doesn't list the units being measured. I'm no gaming geek, but half-life 2 is several years old (released Nov. 2004), and 640x480 is a resolution I haven't heard mentioned in years.
As I say, I'm not a 'hardcore gamer', but I like to play the odd game now and then. But even with my modest needs - a 3-year-old game at low resolution - the GMA 950 would be insufficient. The nVidia chip in the macbook pro would be sufficient, so the technology obviously exists; and I'd happily pay for it. Apple just aren't selling.
In summary people don't like Macbooks' integrated graphics because reviews of that particular hardware indicate it performs poorly.
what's the point of using a PCB these days instead of just putting everything on the same chip? A range of reasons.
1) Apple is not a microchip design company. Hence they don't have the experienced engineers and expensive computer software needed. But between mac computers, monitors, and the ipod they have plenty of experience with PCB design.
2) Some components aren't cost-effective to put on a chip. It's not cost effective to put large capacitors on chip as the die area required would be expensive compared to using an external component.
3) The disassembly shows things like surface acoustic wave filters (where the signal is converted to a compression wave travelling down a quartz crystal). Hard to put on a chip at all, let alone cost effectively.
4) You would have to license the designs for everything you want on your chip. Granted, that's easy with the ARM processor ($1.84 million and the design is yours) but you also need all your radio electronics (which is pretty high tech stuff nowerdays) which might not be offered for licensing.
5) Prototyping with discrete components has negligible cost (Hundreds to thousands of dollars) compared to prototyping microchips (millions of dollars).
6) Different processes. Ever wonder why your PC processor has 4 megabytes of on-die cache and 4 gigabytes of main RAM - why not 4 gigabytes of on-die cache? In short the processor manufacturing process allows more complex, faster designs but costs more per gate. So that 4MB is expensive. RAM costs less per gate by using a cheaper process, making large numbers of gates cost-effective.
7) Likewise you've got a phone camera, which (though silicon-based) requires a different manufacturing process to the processor (it's light-sensitive and has a bayer filter, after all). And you've got a MEMS accelerometer to detect orientation, but that requires yet another silicon manufacturing process (such as deep reactive ion etching (DRIE)).
From the article:
When his third 360 broke, one customer service rep suggested he look into the wiring at his house; electricity problems could have been causing the mess-ups. Problem: none of his other systems (not to mention his several computers and other electronics) have experienced any major problems, and his father is, coincidentally, an electrician. The specific suggestion was brought up by Microsoft customer service again after the eighth console repair. This time, just to be certain, Justin had a contractor come to the house and check the wiring, where he was told that everything was in order, with no abnormalities in voltage of any of house outlets. Nevertheless, customer service has continued to suggest this as a potential cause.
He probably has the thing in a hot spot, like on top of a big CRT monitor, in an enclosed space, in a location with air vents blocked, or next to a hot air vent. We know the XBox 360 has marginal cooling. One would think Microsoft technical support would have taken the customer through questions like 'is it adequately ventilated', 'have you tried using a different power socket', and so on at some point prior to the eleventh replacement unit. I mean, it's a common stereotype of technical support that their first questions are scripted, simple things like 'is it plugged in?'
To me it seems more likely he was sent 10 different refurbished units, and refurbished units have high failure rates.
Or, you know, the story is BS. I mean, this is slashdot after all.
A 20% failure rate of course would not have gone by unnoticed and MS would certainly not have been able to dispute it.
You might be interested in this article: "According to some reports Xbox 360's continue to experience hardware issues. A recent query put to an Australian game retailer puts the figure at a 30% return rate."
However, reading further into the article it isn't very credibly sourced.
Once you start picking subsets of humans who have achieved something, it quickly becomes meaningless.
It's a crazy world we live in - Next thing you know, for Hillary Clinton to become the first female president, or for Barack Obama to become the first African American president, will be considered somehow 'meaningful'.
And that's just the basic mathematic flaw in his reasoning.
I'm not sure I agree with the assumptions you make, but I agree this chap has an analysis that doesn't make much sense.
Consider a 4-option multiple choice test, where you get one point for a correct answer and zero points for an incorrect answer, and there are 100 questions.
0 known + 100 * 1/4 = 25 right 20 known + 80 * 1/4 = 40 right 40 known + 60 * 1/4 = 55 right 60 known + 40 * 1/4 = 70 right 80 known + 20 * 1/4 = 85 right 100 known + 0 * 1/4 = 100 right
Number right = 0.75 * number known + 25
Now, clearly such a test would be BS if the passing grade was 25 right, as everyone would pass. And if the passing grade was close to 25 right (e.g. 27 right) you would get a lot of people passing by luck.
However, if the passing grade is 75% right, you would have to know (75-25)/0.75 = 66.67 answers in order to get a passing grade. And assuming the test designers knew this when choosing the pass mark for the test, they would simply have increased the pass mark to take the relationship into account.
Perhaps the blogger has "35% of people pass" confused with "35% right is a passing grade".
This article isn't very impressive in terms of research, spelling or photographic quality. This is slashdot though, so I guess I can't complain.
When the author says "To be honest, I'm not sure what this machine does", from what I can see of the tiny photo, he's looking at a machine which stencils solder paste onto the exposed pads of the PCB.
When he says "The adhesive needs to be hardened, so the components won't fall off" he means the solder paste is melted then allowed to cool with the components in it, thereby attaching the components to the PCB electrically and mechanically.
When he says "BIOS Taping Area, I'm not quite sure what went on here" I would guess they are writing the BIOS code into the flash memory.
As he doesn't really explain, the reason people are putting connectors on the board manually even after the automated component placement stage is because the plastic connectors would melt in the heat of the oven, before the solder melted. So there are two processes: first the small, high tech chips are put on and soldered in the oven, then people manually insert the funny-shaped easy to melt parts, and they are soldered separately.
Slashdot Mirror
On the other hand, the school, Dwight Morrow High School, shares its campus with "Academies@Englewood": "With more than 400 high-achieving students enrolled, the program --housed in its own building on the Dwight Morrow campus -- offers concentrations in engineering, law and public safety, biomedicine, finance, and information systems." I have to say, I would rather be in the high-achievers school with the decent subjects than the low-achievers school on the same campus which has such a poor selection of subjects.
Just my $0.02.
Having had my apartment robbed, I can tell you that I did call the police. They sent a guy out over 24 hours later, who basically said "yeah, lots of people have been getting robbed around here lately", (note: i lived in a pretty nice area), "you probably won't get anything back. I hope you're insured." and left. No finger printing, no looking at the busted door, no follow up.
I, too, have had several similar interactions with the police - house break ins, car break ins, cars being smashed up, and suchlike.
What you've got to remember is they aren't "the department of useless fingerprinting to make you feel better". Any serious burglar would be wearing gloves. You say you live in a nice area, so it wasn't opportunistic crime, so it's almost certain they had gloves.
And if you think about it it makes sense: A burglar who leaves incriminating evidence at a crime scene and gets jailed for 6 months can only commit two crimes a year. A burglar who leaves no incriminating evidence at a crime scene could easily commit 200 crimes a year. So if you've been burgled, it's 100 times more likely to be the latter burglar than the former.
As I say, I've been in a similar situation; in one instance a guy decided to walk up the street smashing the windows of all the cars he passed. It's frustrating, and He Shouldn't Be Allowed To Get Away With That, but there was nothing to even indicate a suspect, let alone prove their guilt in court. What are the police supposed to do?
There are some things that could be done about this: We could put good quality CCTV cameras watching and recording everything so criminals can be found later; we could fit GPS trackers to all citizens so past movements are logged and can be searched to find suspects; we could put 'a policeman on every corner' literally, so that crime will be seen; we could get rid of probably cause for searches so after burglaries the police can just search the homes of people who have previous convictions, looking for the stolen goods; we could segregate people with a financial incentive to steal into slums which they are not permitted to leave. However, society (justifiably) considers these options unacceptable.
Anyway, here's my point: All the police can achieve once a crime is over is to give you a crime number for your insurer. It's frustrating, but it's reality. Sorry.
On the other hand Cortex-M1 is based on a cut-down version of a silicon core - so there might be scope to cut down the OpenSPARC to a single-threaded single core for FPGA. But at this point I'd be surprised if anyone was running OpenSPARC anywhere except in simulation.
Just my $0.02
p.s. How are you involved with ARM cores on FPGAs at the moment, if you don't mind me asking?
Suppose you instead took the time to find 5 expert, or at least above average, Perl developers at $120k each per year. That seems to be the gist of the article, and it's a pretty reasonable conclusion: Experts can be very much more productive than non-experts.
However, it is also my experience that it isn't always easy to tell highly capable people from the merely capable; that is, I've worked with people who seemed very good initially, but in the fullness of time I realised they were not. And that, of course, is a benefit for having 15 developers instead of five: Any given hiring mistake costs half as much, and reduces your workforce by a fifteenth instead of a fifth.
So how are you supposed to find these expert programmers, and how can you tell a $60k developer from a $120k developer? By asking brain teasers like Microsoft and Google are reputed to do?
Just my $0.02
However, several professions were at one time male-dominated, but have changed with time. One such profession is medicine; see the study Women in hospital medicine in the United Kingdom - in particular this graph. As you can see, in 1960 20-25% of medical school entrants were female; by 1990 50-55% of medical school entrants were female.
Similarly, consider the legal profession. See the (somewhat old) study Women in the legal profession : theory and research, particularly Table 2 (page 3). In 1961 3.9% of Australian law professionals were female; by 1991 this rose to 25% (full time only).
Anyway, here's my point: Some historically male dominated professions have changed to be less male dominated. The same might happen in IT. Furthermore, the inference that current gender distributions imply fundamental gender-dependent abilities may be incorrect.
Of course, IT doesn't enjoy the pay and status of the medical and legal professions, and so may not experience the changes seen in law and medicine.
You would think scheduling a big upgrade for the middle of the holiday season would be asking for trouble.
What's wrong with saying "the second weekend in February" or some similarly random date? It's a weekend so it won't interfere with business, but unlike new years day it won't mess with employees' personal lives too much.
There's a reason businesses and governments don't start their financial/tax years on the first of January, after all.
So what does Dvorak say? The current bubble, already called Bubble 2.0 to mock the Web 2.0 moniker, is harder to pin down insofar as a primary destructive theme is concerned. A number of unique initiatives, however, are in play here. Let's look at a few of the top ideas floating the new bubble.
Neo-social networking. Today everything from YouTube to the local church has a social-networking angle. And this doesn't even consider the actual social-networking sites, from MySpace to LinkedIn to Facebook to even Second Life. This scene is totally out of control and will contribute to the collapse for sure. MySpace was purchased by fox for a somewhat excessive sum, but that's already happened. Facebook's owners reportedly want two billion dollars for the company, but no-one has paid them that. So from this category I see one company, facebook, and it isn't even publicly traded. Video mania. With dozens and dozens of YouTube clones cropping up to get on the "throw money away" bandwagon, you must sense that the eventual shakeout in this space will have a negative impact. Youtube was a rather expensive purchase for Google and it's hard to see where the payback is, but that's already happened. I can't really think of any competitors anyone is likely to invest in... google video, perhaps, but that's owned by Google anyway. User-generated content. This idea has been around since Usenet and just keeps improving. It will make no contribution to the overall collapse except for users reporting the collapse. "This part of the bubble is not part of the bubble"?? Mobile everything. Here is another concept that has been in play since the mid-1990s. It cannot trigger a collapse since it will never fully get off the ground, although the iPhone mania may be a bad sign of something. Mobile what? Are mobile phones a bubble? Or is there a bubble of iPhones and iPhone-like-devices that I'm not aware of? Ad-leveraged search. Most search engines will fail as a matter of course. This segment of the industry is mundane. It would be affected by a crash but not trigger one. You mean Google? Widgets and toolbars. I cannot see the widget scene going crazy, and the jury is still out on toolbars. But there is the potential for nuttiness, I think. The problem here is that these things tend to be dependent on the stability of operating systems and browsers. One bad operating-system patch and suddenly nothing works. There's a "widget and toolbar" bubble? I don't know of any company built around selling "widgets and toolbars".
So, here's Dvorak's bubble of over-valued assets:
MySpace = fox
Facebook - privately owned
Youtube = google
Google
iPhone = apple
Or in other words, the best examples Dvorak has of the bubble of the late 90s repeating itself are:
Fox
Google
Apple
Facebook
Personally I don't agree with Dvorak's assessment that these companies are about to collapse (although it seems unlikely anyone would pay $2 billion for facebook).
Just my $0.02.
2% said they are already running Vista
9% said they planned to roll out Vista in the next three months.
87%, said they would stay with their existing version(s) of Windows.
8% of those polled acknowledged Linux plans and
4% said they would deploy Mac OS X.
I would say "many will stick with the Windows they have", certainly, but I'm not sure I would call 8% or 4% 'many'. And somehow I suspect 'linux plans' might not mean complete replacement of Windows on the desktop.
Just my $0.02
Well, like it or not, a humanities degree is cheap compared to engineering or science. All that lab equipment (and space) costs money, not to mention the people who set it up and keep it running.
True, but surely an engineering department could more easily enter into partnerships with companies, make profit from patents/discoveries/spin-outs, and suchlike?
Humanities departments, you'd think, would mostly get their research funding from the government, because they don't really have a useful product to sell. An engineering school, on the other hand, is probably producing research with a high value.
This doesn't have to mean patents and secrecy; just that if someone discovers something useful, they can work with industry providing experience/expertise in exchange for money. The university/engineer gets money; the company gets a head-start on its competitors; national government collects more tax as companies become more competitive; and humanity as a whole benefits from the new technology getting out ASAP. Sounds like a win-win situation to me.
I'm not sure humanities people can do this much - the latest thoughts on film theory or marxism may be fascinating, but cannot be converted into money in this way.
Just my $0.02.
Furthermore, one would think a device supporting only DisplayPort would command a lower price than one supporting both DisplayPort and DVI, because many people have DVI hardware already, and that price premium would be greater than the licensing fee for DVI.
To me the idea of DisplayPort displacing DVI is similar to the situation with GIF and PNG - namely an established standard with a license fee vs. a less established standard with no license fees*. The outcome of GIF vs PNG was that software gained support for both standards, some people continued to use GIF for backwards compatibility and from habit, and the patents on GIF expired before users stopped wanting GIF support.
I can imagine a similar ending for DisplayPort - namely never displacing DVI enough to save manufacturers any money.
Just my $0.02.
*PNG is a somewhat better standard but has imperfect support - for example Versions of Internet Explorer up to and including 6 do not support native alpha-channel transparency.
Could someone explain to me, preferably without recourse to religious argument, what is wrong with these kids viewing porn?
Nothing inherently, but in the particular case of OLPC I can see why filtering might be reasonable:
1. In a class room situation any web browsing could be disruptive to teaching, but pornography particularly so. It would be pretty weird if at work the guy opposite me in the office was looking at porn all the time!
2. Parents may be reluctant to give children access to OLPC machines if the machines have a reputation as 'porn portals'. Non-adoption would obviously prevent the anticipated educational benefits of OLPC being realised. To put it another way, revolutionising religious views of pornography is not part of OLPC's core aims.
3. Similarly, if looking for government funding in the US, it's probably useful not to have the stigma of pornography hanging over your head - what with all the religion involved in US politics etc.
Just my $0.02
If I look over your garden fence and see that you're building a giant widget and then you notice and offer me a tour of your giant widget do you have any legal recourse if I decide that I like the idea of having my own giant widget and then make one for myself?
:-)
It sounds like you might have just deduced why legal constructs like 'patents' exist in many countries
Furthermore, if you write to television licensing and tell them your TV is used for gaming/DVDs only they will stop sending you letters; and if an inspector visits and you show them the TV connected to the games console, they will say "that's fine, you don't need a license". I have verified the two preceding facts by direct experience at two different houses within the last five years.
1. Park my bike, shower etc.
2. Turn on computer.
3. While it starts, get a coffee.
4. Log in, drink coffee, check e-mail/calendar.
5. Get to work.
I've got to say, that sounds like the sort of interview question that would get some pretty boring responses. Like mine, above. So I usually jazz it up a bit in interview:
1. Park my unicycle, change out of my superhero unitard.
2. Get a new guitar from the IT guys because I smashed mine at the end of my last performance.
3. Check in with each of the 10,000 people who work under my command, all of whom I know by name.
4. Have my executive assistant relay my e-mails to me, one character at a time, by throwing lettered frisbees back and forth between my company's two tower blocks.
5. Take my second breath of the day.
So far I haven't had any job offers, but I figure the market is pretty competitive at the moment - it's only a matter of time!
As far as I can tell that first review says at 640x480 you get 6.4 frames per second in Half-Life 2. The second review lists the GMA 950 as not performing very impressively, though it doesn't list the units being measured. I'm no gaming geek, but half-life 2 is several years old (released Nov. 2004), and 640x480 is a resolution I haven't heard mentioned in years.
As I say, I'm not a 'hardcore gamer', but I like to play the odd game now and then. But even with my modest needs - a 3-year-old game at low resolution - the GMA 950 would be insufficient. The nVidia chip in the macbook pro would be sufficient, so the technology obviously exists; and I'd happily pay for it. Apple just aren't selling.
In summary people don't like Macbooks' integrated graphics because reviews of that particular hardware indicate it performs poorly.
1) Apple is not a microchip design company. Hence they don't have the experienced engineers and expensive computer software needed. But between mac computers, monitors, and the ipod they have plenty of experience with PCB design.
2) Some components aren't cost-effective to put on a chip. It's not cost effective to put large capacitors on chip as the die area required would be expensive compared to using an external component.
3) The disassembly shows things like surface acoustic wave filters (where the signal is converted to a compression wave travelling down a quartz crystal). Hard to put on a chip at all, let alone cost effectively.
4) You would have to license the designs for everything you want on your chip. Granted, that's easy with the ARM processor ($1.84 million and the design is yours) but you also need all your radio electronics (which is pretty high tech stuff nowerdays) which might not be offered for licensing.
5) Prototyping with discrete components has negligible cost (Hundreds to thousands of dollars) compared to prototyping microchips (millions of dollars).
6) Different processes. Ever wonder why your PC processor has 4 megabytes of on-die cache and 4 gigabytes of main RAM - why not 4 gigabytes of on-die cache? In short the processor manufacturing process allows more complex, faster designs but costs more per gate. So that 4MB is expensive. RAM costs less per gate by using a cheaper process, making large numbers of gates cost-effective.
7) Likewise you've got a phone camera, which (though silicon-based) requires a different manufacturing process to the processor (it's light-sensitive and has a bayer filter, after all). And you've got a MEMS accelerometer to detect orientation, but that requires yet another silicon manufacturing process (such as deep reactive ion etching (DRIE)).
That's all I can think of for now.
To me it seems more likely he was sent 10 different refurbished units, and refurbished units have high failure rates.
Or, you know, the story is BS. I mean, this is slashdot after all.
A 20% failure rate of course would not have gone by unnoticed and MS would certainly not have been able to dispute it.
You might be interested in this article: "According to some reports Xbox 360's continue to experience hardware issues. A recent query put to an Australian game retailer puts the figure at a 30% return rate."
However, reading further into the article it isn't very credibly sourced.
Once you start picking subsets of humans who have achieved something, it quickly becomes meaningless.
It's a crazy world we live in - Next thing you know, for Hillary Clinton to become the first female president, or for Barack Obama to become the first African American president, will be considered somehow 'meaningful'.
And that's just the basic mathematic flaw in his reasoning.
I'm not sure I agree with the assumptions you make, but I agree this chap has an analysis that doesn't make much sense.
Consider a 4-option multiple choice test, where you get one point for a correct answer and zero points for an incorrect answer, and there are 100 questions.
0 known + 100 * 1/4 = 25 right
20 known + 80 * 1/4 = 40 right
40 known + 60 * 1/4 = 55 right
60 known + 40 * 1/4 = 70 right
80 known + 20 * 1/4 = 85 right
100 known + 0 * 1/4 = 100 right
Number right = 0.75 * number known + 25
Now, clearly such a test would be BS if the passing grade was 25 right, as everyone would pass. And if the passing grade was close to 25 right (e.g. 27 right) you would get a lot of people passing by luck.
However, if the passing grade is 75% right, you would have to know (75-25)/0.75 = 66.67 answers in order to get a passing grade. And assuming the test designers knew this when choosing the pass mark for the test, they would simply have increased the pass mark to take the relationship into account.
Perhaps the blogger has "35% of people pass" confused with "35% right is a passing grade".
This article isn't very impressive in terms of research, spelling or photographic quality. This is slashdot though, so I guess I can't complain.
When the author says "To be honest, I'm not sure what this machine does", from what I can see of the tiny photo, he's looking at a machine which stencils solder paste onto the exposed pads of the PCB.
When he says "The adhesive needs to be hardened, so the components won't fall off" he means the solder paste is melted then allowed to cool with the components in it, thereby attaching the components to the PCB electrically and mechanically.
When he says "BIOS Taping Area, I'm not quite sure what went on here" I would guess they are writing the BIOS code into the flash memory.
As he doesn't really explain, the reason people are putting connectors on the board manually even after the automated component placement stage is because the plastic connectors would melt in the heat of the oven, before the solder melted. So there are two processes: first the small, high tech chips are put on and soldered in the oven, then people manually insert the funny-shaped easy to melt parts, and they are soldered separately.
And now you know!