If anything, AMD is shooting itself in the foot by betting the entire farm on a 64-bit archtecture that will have no software support for years to come.
If they can maintain a 1:1 performance on existing code, then the extra price-tag should be warrented for people that actually need it.
If the costs are no more than 1.5:1 for the 64bit CPU, then they can gut the IA-32 only line. Future die-shrinks should bring down costs of the 64bit chip and make it close to 1.1:1 cost, at which point they'll be posed to make 64bit their main line. Marketing wise, they'll be able to say, "we have a chip comparible in price/performance to intel, BUT we have 64bits".
The danger, of course is that any speed-ups applied to the 64bit archetecture (for 32bit code) can be applied to their athlon line (or it's eventual 32bit replacement). Thus they can always produce faster/cheaper IA32 chips, hurting the AMD64 demand.
The only reason to market AMD-64 to the desktop is to try and hurt Intel (priming the market and setting a standard). But they would have to forego lucrative "server" prices.
The biggest thing that got me started down this path was not only the 4gig limit but the fact that the PIII changed form factors so many times.
It's actually funny because it's generally rare to be able to apply newer chips into older motherboards. In other words, this is a solution that has an artifical problem. That PC chips were so cheap and easily available that it was worth maintaing a rapidly outdating sub-system (which comprised an increasily important role in overall performance).
I know several non-PC vendors attempted to use CPU-cards, where the CPU/cache subsystem were a mini-motherboard, to minimize the upgrade pains. Interfaces between sub-boards were standardized, etc. Problem is of course: 1) added MB cost 2) radical architecture changes require upgrading the interface (Pentium II and up can be thought of having this problem since CPU+Cache+MMU were on one mini-board) 3) anti-marketing... Why buy a chip when you can buy a whole new computer (and resell your old computer intact).
I actually gave up on the x86 CPU upgrade path a while ago when I started on Linux.. Instead of spending good money on bad, I kept the aging PC on the network but removed it's monitor.. Strapped Linux on it and played with Beowolf. I've owned several dozen machines in my day (propably purchased a couple dozen more CPU's for my early days of CPU-upgrades).
Segment registers are quite useful if you're doing any real OS development.
Except that it's highly non-portable. The types of optimizations you can make with segmented memory spaces generally affect the entire operating environment. Theoretically, the OS could provide hooks that return not only an address, but the associated segment register. Now you have entire applications requiring segmentation. (Doesn't win-16 code do this? I only ask because I see segment-register resource counts in Norton System Info.)
I believe the IA-32 segmentation representation utilized addition to produce physical addresses. While this allowed arbritrarily sized and placed segments, it introduces a performance hit. It's been a long time, and correct me if I'm wrong, but I believe the VM paging system is stuck with addition instead of "and"ing as well. So there is a very real cost that came about from this delayed transition to flat memory spaces.
IP-ID is not for reconstruction but identification.. It's like a database RowID.. Sequentiality has no bearing except that auto-incrementation is easier to implement than true random number uniqueness. mssql and other databases, however actually do perform random number generation (checking against a database of previously used values).
In any case, IP-ID is good for ICMP requests (such as ping)... If someone drops your ping's echo-request packet, they can be nice and send a response saying I've gone and dropped your IP package uniquely named IP-ID. Many servers don't do this anymore for fear of the "ping-of-death" DOS attack.
As for your question of how to reassemble packets. This is only really relavent for non TCP protocols (ICMP, UDP, etc). TCP has it's own sequence counter and can easily facilitate mini-sized IP packets. UDP can theoretically send 64k messages however and relies apon the underlying IP to fragment.
My memory is a bit fuzzy, but there are additional IP fields that specify which fragment out of n-total a given packet is. The idea is that routers/gateways can reconstruct/break-apart the packet arbitrarily at each leg of the internet's journey. I believe that each fragment still carries the originating IP-ID.. But it is the IP-ID + fragment-number + timestamp that uniquely identifies a packet (since a server will eventually reuse old IP-ID's).
The problem with filtering is that most people have "good mail" go into their inbox and spam mail go into a spam folder.. This is ludicrous for several reasons.. What if you're on a mailing list that sends dozens of emails a day (perl-6, for example). What if you miss who the sender is and you make assumptions based on the subject line or header, only to draw incorrect conclusions.
The answers my friends are to folderize your emails. As best as possible EACH sender should have their own inbox. This removes the subtle issue of temporary mistaken identity (since you have to physically navigate to their folder). More-over, for spam addresses, they can easily be flagged as spam and go into a sub-category that easily gets closed off.
One issue, however is the priority of filters (ordering for evolution/kmail/etc). If you use 3'rd party spam filters, make sure that the very first action is to weed in recognized addresses. This avoids your particular problem. The only caveat is that some viruses are sent from trusted senders. Thus any virus checking (which here is only currently relavent to windows [l]users) should go to a sender.virus folder, or rename the subject line to VIRUS: $orig_subject.
Unfortunately company-wide virus checkers are difficult to reconfigure in this manner. At our site, we do the subject-rename and let the email pass through. Thus our workstations can easily handle such rules however we choose.
While I don't know about the Rubidium-80 reference, the difference is quite possibly the poly-exclusion principle. I have only casual knowledge in the topic, but it seems to me that interaction-based quantum entanglement would involve time-space-state cohabitation which fermions generally can not do (by definition).
Thus, if you can transport the state (spin, charge, etc) of an atom and impose it on another atom, you have effectly transported the atom, even though no mass travelled the distance.
Except that this isn't necessarily useful with respect to either faster-than-light travel or moving massive bodies about, since quantum entanglement requires initial contact as far as I understand.
It is probably useful for secure communication since you initiate a beam in the middle going in opposite directions, then manipuate the photons at one end, and read the photons at the other (I realize it isn't this simple). Here it isn't the bandwidth, but the state of the photons that are of interest.
> cmd.exe? I think the point was that windows doesn't facilitate command-line friendly applications (You're more encouraged to do GUI and only GUI).
Therefore, no shell can compensate for the lack of glue that application developers provide.
It does seem that MS is on the right track though. Assuming that administrators are willing to learn the ins and outs of VB.NET, they'll have access to the entire.NET API of an application. Essentially administrators could be programmers at a lesser level (only having to learn the very public APIs (such as print, convert image attributes, etc) similar to OLE features). It's much more powerful since application developers don't have to think about what to export; if they write inter-application API's, it'll be available to the sys-admin.
Course the same thing already exists with java, but there just isn't enough java in OS's tool-boxes to really be considered here.
Ethical problems don't exist... It's a made up thing, a simple case of "I believe this, so I will force it upon you".
Only if you're a firm believer in chaos. While I'm not a sociologist, it seems to me that laws are enforcements of local social norms.
Obviously we don't put you in jail for masterbating (well, in private at least), yet that can be argued as mass genocide (an entire potential sub-race). On the other end, we tend to get upset when a 1 year old is tortured to death.
The gray line in between can only be drawn by such social norms. The only thing I know for sure is that ANY such lines should not be uniform. Different regions of the world are composed of different biases, and thus would give you different stopping points of acceptability. So long as we don't have a single common law for such limits, then I think the world will be fine.
The grand unification (aka federalism / globalization) stands to get in the way.
Part of the obvious problem here is the bias to-wards a soul. Some believe the Soul is introduced at conception, while others at first sign of activity. Most of us probably treat a baby like a chicken-McNugget; We eat the McNugget so long as we don't recognize the dead cow in its shape/smell/color. We're happily ignorant of the horrible life-style of said cow. Taking the life of something that doesn't "feel" human is easier to avoid guilt/shame. Of course, still others don't believe in a soul, so the problem becomes a practical one.
In any case, the "Ethics" do in fact exist, but in a subjective form. This doesn't diminish their importance in decision making.
Personally I believe all living cells share the same life-force; whatever that might consist of (be-it an elaborate bio-electronic circuit, or some netherworld spirituality). It's merely human bias that places human life over that of a mouse (a bias that's justifiable, since other higher sentient forms will also sympathize/defend their own). A human cell in a mouse's body is still a human cell, and still alive. Human DNA in another being's body isn't much different. The trick; I think, is that when you have a sentient being, then you remove it's sentience, then it will experience death (loss of sentience). To what-ever degree we can determine that threshold, I would feel would draws the line of morality. Otherwise, the individual cells that make up a pre-child are nothing more than worker cells; such as blood-cells, hair follicles, etc. While their life is precious, they have little value atop that of saving the lives of thousands though brute-force disease research.
There are those that feel that stem-cell research and aborted fetus based research will encourage abortion-for-profit. Thus, even if the subjective matter of non-potential-life tissue isn't strong enough, they'll claim secondary immoralities. My response is that regulation can easily be applied to restrict such actions. Social norms will undoubtedly prevent mass-human synthesis (at least for the time being). Even if the future holds some perverse sub-human programs, I'm sure that our child-like incompetence of the intricate workings biology will facilitate normalizing natural disasters (such as the synthesis of new diseases, or any other sci-fi outcome).
I agree with what you're saying.. And I was in no way putting down literatures et. al. I was merely saying that I was never inspired in those fields, and so I never excelled. Thereby being evidence of the importance inspiration (given my attributes in other inspired fields).
More-over, I never said to home school. I merely said that teaching has a definite limit; you can bring a horse to water, but you can't make him drink. I guess I neglected to add the value of a mentor; someone that can answer frustrating questions.
The best way I can summarize my intentions is to say to simply be a role model for your children. If they want what you are and have, then they may have inspiration to follow in your tracks and take on the passions that you have. If your passions are TV, video games and socializing, then you can't get too upset when your children don't do well in math. If you obsess over software programming and don't portray a practical benifit to them, they may simply ignore you; writing you off as an eccentric.
As a youth, my father could work wonders with electronics and cars. We had propaganda about how great Einstein and Benjamin Franklin were and what they were able to do in their lives. There were the stories of rags-to-riches by various people throughout American history. These things provided me inspiration; and being an introvert, I wasn't AS distracted by the social aspects (though there was a lot of TV). My point is that it's a delicate balance. Tiny things can influence monumentally for better or worse.
I guess one active thing a parent can do is to keep ready track of who their children's mentors/idols are. They're generally open about them/ willing to discuss them. Course, when they fall in love with a rock-star or other such hard-to-reconsile idol, I don't know what one can do. But as you said, you have to let them be their own person.
If you really want to edumicate your kids, do it yourself.
Actually, while I don't know what kind of studies there are for it, the best way of "having a child learn", is when they learn for themselves.
Teachers necessarily can't know how to talk to a brick wall effectively, yet that's exactly what we bring generation after generation up as.
In com sci classes at least, the people that require tutors, generally don't do well; requiring spoon-feeding and "can't you just do my project for me". Those that get deep into the material on their own can often ace at least introductory course-work (com sci can definitely be a mind bender, especially when dealing with number theory).
The key is motivation. If you are unmotivated when you meet a person, you may remember little about them. Names are most famous: Many people (myself included) are already pessimistic about being able to remember hair/eye-color, names, etc. so I don't bother paying much attention when I meet someone.. Sure enough, it's a self fullfilling prophesy.
People going to computer or math classes with such pessimism have little chance of succeeding (regardless of their background deficiencies).
In my life, I've found that having desirable projects that happen to require learning a lot about a given topic affords an ample amount of motivation. In science, I crave sci-fi concepts; I want to understand them so I could possibly invent something new. With computers, I develop overwhelmingly complex goals (on the MRPG scale). Thus virtually every aspect of science, math and computer skills have been on my "I need to know" list.
Conversely, I haven't found such motivators for history, art, music, literature, so I only give those subjects a necessarily passing glance. (Though at some point I developed an appreciation for the story-telling nature of history).
While being totally non scientific (effective sample pool of 1), I still see such trends, and believe that inspiring your children in the single best way to teach them.
The trick is of course, how to inspire. And how do you avoid making a project obviously contrived to the point of frustration.
I'm afraid not. E-Mail allows me to send a message, or respond when I want to. Much better flexibility than IM.
Actually, when I used ICQ, I admired it's treatment of messages as mini-emails. If you were offline when you got a message, it would be available for you when you logged back in.
Therefore, it's perfect for sending offline important messenges that need greater priority than spam-neighbored emails (which people classicaly think to check periodically instead of continuously).
Effectively, ICQ was equivalent to an email client with a heirarchy of per-sender mail-boxes, where only the most activly recieved are up front (such as a spline tree). If you could set the "you've-got-mail" equivalent-tone to only activate when a top tier (say 10 senders) give you new mail, then you'd effectively have the same thing, though for high-volumen, it wouldn't be as efficient (due to TCP session per message-group, and header over-head).
Summary is that there's evidence of a holographic aspect to neural cognition, and thus raw TFlops are hard pressed to simulate it's computational complexity. Thus the brain is still far ahead of computers by a couple orders of magnitude.
Also, I would split them "vertically", not "horizontally": multiple competitors that each produce the OS and applications, not a single company for each product.
I'm sure this topic has been beaten to death, but I'm not quite sure how this would work (haven't thought about this in what must have been years).
Either: A) give each fork a complete licence to the existing code, thereby forking all the software B) mix and match segments, effectively doing both vertical and horizontal splits. (One doing MS Hardware and some but not all office products, but both doing Office/Windows (the main point of contention))
If A, then both companies would simply duke it out until one won (namely dirty incompatibility tricks, having even less incentive to share APIs). Note that winning=loss of profitability for one company.
Note if B, then you don't avoid the problems of A, but you tend back towards verticle non-competing companies.
I'm just not invisioning how horizontal could work.. With AT&T, you at least could regionalize the market and then regulate the leasing of regional lines to non-regional companies. But we're talking about a software commodity (at least at first).
You're talking about the substitution effect. The previous guy was talking about Supply and Demand.
There is a certain demand for PC's. Some people can NOT afford a $2000 PC, no matter what the configuration of costs is. Many people can afford a $150 console gaming system, however. Likewise the demand for consoles is higher than for PCs. To some people, a MAC or Linux are not substitutes. Thus, they may simply elect to not make a purchase (this includes deciding to not purchase an ADDITIONAL PC for say their children).
Thus, you have a demand curve that's directly related to the price. Higher price means fewer purchased PCs. While there is a substitution effect, it is completely practical to say it is only a small factor (given the realistic monopoly of MS on our everyday computing needs).
Thus MS, being a veritable monopoly on a virtually zero marginal cost product (e.g. software) can greatly determine the price.
However, If they charged $1k for their OS, there would be fewer purchasers. Revenue = quantity_sold * unit_cost. At some point revenue will peek for a given price. The propensity for consumers to pay higher prices without losing too many sales is the elasticity curve. You can be absolutely sure that MS is keeping VERY close tabs on the elasticity of their product, and charging accordingly.
Further, they are working VERY hard to artificially enhance the elasticity curve. By regularly making OS releases, there is an irrational desire to "have the latest and greatest". Irrational only in the sence that experience should teach them that MS hasn't provided many good "new releases".. They've often been buggy and filled with "more of the same". Still some are attracted to the glitter of new graphical UI's, and they justifiably acquire some utility; though it is doubtful that that utility equals the cost of an upgrade.
MS goes one further by obsoleting existing software lines (due to non-backward compatibility). New games require new Direct X engines, which aren't provided on successively older OS versions. Office is necessarly non backwardly compatible. MS seeds new markets which will generate interest in new software paradigms (such as.NET) which will of course require upgrading.
All this equates to maintaing a heafty elasticity and demand curve. This more than compensates for the substitution effect as it exists today.
There is one good thing to come from all of this.. As they price themselves closer and closer to their desired elastically justified point (held back only by sticker shock psychlogy, and fear that the government will intervene), they lower the "barrier-to-entry" costs that were fundamental to their anti-trust-trial. It becomes ever cheaper relatively-speaking for a company to choose Linux. Moreover, it becomes more viable for a 3'rd party to charge money for a newly developed UI for Linux. Moreover, companies (such as Win4Lin, and others) are already doing it.
The only obsticles are compatibility.. But such a commercial company is now legally required to have access to such compatibility information from MS (as far as I understand the most recent rulings).
Unfortunately, MS knows this, so their cost increases will probably continue to stagnate; being just above what is acceptible, but not above that critical barrier to entry.
The article's paper is jiberish in the sence that they are merely demonstrating what we already know.. MS is not constrained in terms of pricing by competition. I was also upset to read what MS "could" have charged, since this neglects the billions spent in R&D.
The problem is that Windows currently takes the role of stanrdard oil or AT&T.. A necessary comodity that is being charged well beyond the reach of needy people. Walmart is selling $200 PCs ($300 if you want windows). That's 50% of the cost for a barely useful machine (web browsing and that's it). To write a term paper, they'd have to shell out the equivalent of 200% of the cost of the machine. The paper is obviously dated in this respect. What we really need is a paper on the effects of harm to the public due to excessive charges.
The OS itself is only worth maybe $25-$45 dollars, but MS doesn't just sell an OS. They provide lots of "innovation" for additional fixed-cost. The monopoly argument must be tied to the harm of the population (as with AT&T / standard Oil) for our corporate friendly Republican congress to give this issue a second glance.
Monopoly is the ugly boil of capitalism that nobody wants to admit is real.... According to contemporary economic theory, profit only occurs where a monopoly exists.
While I've minored in economics (and don't pretend to be an expert), I have a few beefs.
First, due to diseconomies of scale, it's entirely possible to regionalize a product. Franchises are somewhat related. Independent ownerships (and pricing) with royaltiest paid. Technically, McDonalds wouldn't be a monopoly, so long as it was regionalized.
Moreover, I've always hated the concept of zero profit. There is a tremendous false impression that zero profit means no money is made. The simple fact is that executives earn a salary, and that salary is counted as part of the fixed and or variable costs (variable if they charge hourly (which counts if they are compensated via bonuses)).
In a perfectly competative market, theoretically any possible reduction of variable costs lowers the market price. Thus, companies with lower executive pay-offs would affect such a change (given the greater proportion of their salary v.s. a workers salary). In reality, however, there are few truely "perfectly competative" markets. This is polluted from things like brand-name recognition, or similarly risk-aversion costs (due to not purchasing from known name brands), or locality/availability. Computer fairs, for example would be a good candidate for perfectly competative markets, but several dealers are shady, and those vendors set up next to the front door are highly likely to charge more for the same product.
The fact that many such computer companies are family owned (possibly even with family connections overseas for lowered costs), the employee costs are directly tied to firm profits. Yet, obviously if there was zero increase in price from OEM purchase to customer sale, the companies would all go out of business. Yet, somehow it's still worth the while of hundreds of companies to visit these computer fairs.
The reasoning is that everyone must generate at least a penny of revenue in excess of hardened marginal costs (meaning raw materials). That penny necessarily goes to the variable cost of labor. But that labor is directly tied to "profit" of the owner. It is then the goal of marketing / business strategy to command greater than 1 penny "profits".
The point here was merely debunk the notion of zero profit and perfect competition. Again, technically all labor expenses should be summed into the profit category since they are variable and expendable in direct proportion to the level of competition.
Beyond that, I believe that the concept of Average cost and short/medium run plays heavily into the practicalities of profit maximization. Theoretically in the long run, Average cost equals Marginal cost, and therefore price = MC due to competition. From this, only Monopolies can maintain a profit. However, no product/consumer is static. Virtually every product has evoked change in supply or demand to a degree that warrants both newer R&D costs (contributing to higher Average costs) and to short-term niche markets which allow for higher consumer elasticities.
I am thinking of the entry to maret as a diffusion rate with a natural impedence (resistive) force. The rate of change of a market is directly proportional to the ability to command profits, since there will be more opportunities to command a niche with little competition.
Further, given the set of markets that are changing (arguably a majority), and that AC = MC only after an infinite time, then for this set, AC != MC necessarily. Thus the dynamics of such transient markets are too complex to generalize. For some firms, costs will be lower for market entry, and thus AC for them will be lower than for competition. However, that market will eventually saturate or be obsoleted. At that point, transition can occur where the process begins again.
Such a system is highly susceptible to marketing (information scarcity / irrational purchasing), assymetric advantages (family connections, illicit collusion, etc), and simple luck (being at the right place at the right time at the right price with the right product).
My point in all the above is merely that the rate of change of a market is sufficient to allow for non monopolistic / non-perfectly-competative markets that could very well have been either. More-over, the food industry / computer industry necessarily falls into this category of rapid change (new "better" products replacing older ones).
I'm sure a lot of research has gone into this area, and I freely admit that I'm out of my leage. I'm simply [trying to] debunk what seems to be the elementary/classic view. You "can" make a "profit" from your hard work; you simply have to get past the idea that your dividends from your IPO'd company are the be-all-end all (completely ignoring the lunacy of capital gains off IPO'd shares, which is nothing more than gambling). (And yes I realize that economic profit covers more than dividends, but it is merely a different scale).
My main motivation for the above rant is the notion of fixed wages or worse yet, salaries w/in a corporation. Company effibility (to outside investors / analyzers) ignores the economic welfare due to employeeing potentially thousands of people with very high salaries/wages. If, instead, there were low salaries and instead a heavy bonus structure directly tied to contributed value to the company (i.e. productivity), then expenses would be more variable, and the need disperse profits would deminish; bonuses could be proportional to profits, and annual reported losses would be significantly reduced. The system rewards productivity retroactively instead of proactively. Mistakes / setbacks are more quickly regulated in terms of firm performance-analysis and future decision making. Most notably, there are fewer sunk costs.
I'm ranting at 2am, so I appologize. The more general argument, however is the viability of smaller, regional firms with little or no stock-based control / revenue streaming. The goals are a reduction in what I consider the adverse wants/effects of corporate economies of scale (namely corporate immorality/ neglect/devaluing of consumers) which is a natural extension of the unregulatable masses of businesses in a market based economy.
They charge you $1.25 for seventeen cents of syrup and some essentially free carbonated water.
While it's true that soda has a high elasticity (and hense viability for high profit margins), this is only a tiny fraction of the cost of a restaurant. McDonalds isn't exactly doing well these days, and I'm seeing more and more restaurants charge higher and higher prices for their soda (fountain or not). The issue is that while the marginal costs for McDonalds are very very low (cheap labor, mass produced fries, burgers, soda), the Average costs are very high (economic-rent, expansion costs, pyramidal corporate heirarchy costs, dividend payouts, etc). McDonalds is only really cash positive during their rush hours. But so long as they're open during slow periods of the day, they're losing money.
Like any rational being, they are maximizing the revenue based on elasticity curves.. And the fact that restaurants as a whole charge fabulously for soda reduces the substitution effect and thereby augments the elasticity.
Note I said reduces, and not eliminates.
Though I have grown completely tired of McDonalds, I'm still a poor college student, so what I do is buy bulk Soda from Grocery stores while they're on sale (I'm too poor to even join a Sam's club). Then I go to McDonalds and purchase from their Dollar menu, and bring the food home and drink my bulk soda at $0.25 / can. The net effect is that I take advantage of their speedy service, decent volume/taste and most imorptantly, regionality (one located every so many miles).
Unfortunately McDonalds does not reap the benifits from people like me. I generally acquire a great deal of value from my purchase, and most likely press up against their marginal costs, to say nothing of their average costs. I can see the effect as they slowly discontinue items from their dollar menu (such as their double cheese burger; arguably, their greatest value for some time).
I know, it's faulty logic on MS's part; I recently bought licenses to all my illegal MS software because the university was selling them cheap. Before, I couldn't afford office and windows XP and vis-studio.net, so I stole them. Then I paid about 50 bucks and got licenses.
MS knows what they're doing. While they lose SOME money to piracy, what they ARE doing is using scare tactics to ensure compliance from businesses, who command fabulous elasticities. A couple hundred dollars per employee is nothing to a business; especially considering the cost of Office furnature (even a computer these days is trivial; I should know, we run a small business, and have been plagued with such growth costs). The risk of being audited (these days) has a much higher associated cost (especially when factoring in the bad publicity to clients), and therefore it's physically cheaper to maintain compliance.
This goes for both Office and the OS, which is why it's harder and harder to even be able to afford MS Access (being bundled with ever more expensive packages). The problem to people like me is that I am made less valuable when consulting for friends / businesses by not having affordable access to the latest technology (while they (being part of a compliant, modern company) are generally limited to non-backward compatible versions).
I have absolutely no problem with MS attempting to copy-protect their own works (via online registration, etc.). This issue makes it exceedingly difficult to attain educational copies of Office / MS SQL, etc. Note that I'm not kidding about the word educational. I don't use MS for any personal activities, nor contracted activities.. However, being a software developer, I feel the need to compare/contrast and consult on MS technologies. Their "student editions" are crippled and bare little resemblance to their production products. Therefore, the only way to "learn" their material is to outright purchase full copies. This is of course unacceptible for such marginal use (e.g. it would be cheaper for me to purchase linux HW for my perspective clients than to purchase the latest MS SQL to lear how to help a friend setup such a machine).
However, I am actually one of MS's targeted platforms. In order to sell myself, presumably, I must be a MS developer (paying $1k / year to MS), and I must regularly certify myself and thereby take certification classes. Given that I feel that MS products are inferior in general (though there are certain superiorir attributes, relating to UIs), I do not wish to lock myself in. I choose UNIX development (Solaris/Linux/IBM, etc), but sadly this is not fully practial.
Thus, like you, I will sorely miss the ability to acquire evaluation copies.
While it is surely possible to "crack" most anything. The legal ramifications are becoming increasingly costly. Moreover, as I've repeatedly learned, cracked copies may contain maliscious code or may simply be crippled in unforseen ways. My girlfriend was unable to install Photoshop among other products on a cracked version of XP. Eventually I recommended purchasing an OEM copy and life was better (especially considering lowered tech support issues).
The end result of this disertation is that Nash/Adams is upheld here. We are right in secrely pirating (meaning we must be dignified in accepting defeat if we are caught), and MS is right in thwarting our piracy. A system where you can't cheat is fundamentally better than one where you shouldn't cheat.
Well put. Just a few additions. A capitalistic society is necessarily a buyer-beware society. One of adverse selection, and price/quality volitility.
However, it is also a game of rationality. With the Adam Smith (society benifits from each acting according to his best interests), or better John Nashes addition of collective barganing, we can compensate for corporate abuses by: 1) Having a reliable system of regulatory law 2) Having an adaptable set of regulations
Part 2 is qualitative at best, but generally should work for major issues (supposedly MS should have been such a case).
Part 1 is the most important. In developing nations, it's less likely to exist. Mafia, corruption, beuracracy, and fear of hurting the economy by punishing opportunistic yet successful businesses all frustrate this goal. However, even after successfully enforcing laws, it is important to make sure that there is a very low probability that a corporation can "get away with" violations. Lacking such accountability, corporations can factor in such prosecution into the cost of risk. Prosecution is no more coercing to them than, say, the risk of losing a couple shipments of supplies/product due to bad weather.
Prosecution should be nothing less than detremental to a business, and there should be no [risk of] leanency(sp?), for fear that the process can be treated as insurable loss. Arthur Anderson is a wonderful example of such a practice. It'll be another decade before an accounting company would risk such data-munging.
This is the only way that a mostly free capitalistic society can prosper.
Personally I feel that elements such as software patents are egregious, but these fall into the qualitiative elements of regulation. So long as item 2 is upheld, however, the long term should not be the worse for it all.
Well, and breaking up big monopolies is a good first step towards addressing that problem and restoring a free market.
While I'd love to see MS broken up; I highly doubt that it would solve much. It would only get deregulated, given enough time. MS and Intel are in collaboration (due to mutual advantage). Why would OS / Office companies be any different.
Office, for example, proportedly would better support Apple / Linux. BUT, Linux support would cost them money initialy (and would require nightmarish tech-support), and would open up the Linux platform. Doing so leaves them vulnerable to competition, such as from Star Office. Further, I believe many proposals had share holders owning stock in all sub-divisions. They would obviously further the collusion efforts in any ways possible.
Remember, we're living in the same as as Enron, with accounting and inter-corporate cohabitation that puts even MS to shame. Any circumvention is possible.
Lastly, no government is going to break up MS any time soon; at least not until we're out of a ression. MS is a large contributor to our international exports; thereby lessoning the trade-deficit.
I remember seeing one billion dollars thrown around several times for the advertising budget.
I might be off, but I remember a $1B advertising campaing related some a MS product recently; might have been WinXP. The analysis of it was that this statement was pure marketing.. They included in that money expenditures of OEMs, and other creative aspects. That particular campaign was more like $200 Million physical MS dollars.
Given the ruthless rational marketing gaming they're playing, I wouldn't trust any numbers that come from MS. Remember that they have to justify their [low] expenses, not only to the government/public, but to their share holders; many of whom have been crying for years to have dividends paid out.
Actually if you're smart about it, you don't have to deal with spam if you're willing to risk losing mail.
Simply treat your inbox as a pseudo-junk-mail folder. Basically repeated spam gets filter rules to go to junk-mail/be deleted (I'm paranoid and never delete mail; disk space is cheap). Periodically scan the inbox for new contacts which should get their own personalized folder+filter, and scan for new bulk contact domains to add a junk-mail-filter.
I get between 25-200 messages a day, 90% junk-mail, and I never have a problem corresponding with close friends.. Some times a rarely-talked-with associate will have to wait a week before I scan through the inbox to find him as non-junk-mail, but I find that acceptable.
Go to any online computer vendor (Not sure about Dell), and notice the purchase of Win XP Home Edition at around $89 OEM. Vendors that perform bulk transactions may have special secret deals that cost them less, but such vendors are notorus for charging 100-1000% profit-margins on accessories (such as for 128Meg of RAM). Built a better custom machine for a friend than Dell was offering for about 2/3 the price, and that INCLUDED XP Home pre-installed. The only difference is the potentially extended warrenty offered by Dell (and of course, brand name recognition). I even give thumbs down to Dell and friends web site, since they don't give up front pricing on components (mainly package deals).
Anyway, point is, unless you are limiting your options (e.g. Windows-aligned dealers), then Windows isn't a sunk cost.
If SRAM were as cheap as DRAM we'd be using it for system memory and might not even need cache at all.
Not necessarily. There's an inherent slow-down associated with large address spaces. Not to mention the heat decipation. Heck, why else do we have 3 to 5 layers caching? The practical approach is to have successive layers of cheaper, larger and slower memory.
Since we already have 8 meg caches (in some high end machines), there's little value in doing away with multi-gig low-power, low-cost memories. Theoretically some apps will achieve noticable performance gains, but at enormous costs (today at least).
Furthermore, DRAM with internally managed refresh logic is functionally identical to SRAM (but non-deterministically slower). For something like video memory which regularly touches every byte of memory, the refresh logic would be unnecessary; thereby speeding up the memory. Further, DRAM is sufficiently performant enough to handle refreshes. 4MB * 80fps (for true color 1280x1024) = 320MBps. DDR can handle 2.1GBps alone. This doesn't even acknowledge the possibility of interleaving/banking/segmentation or what-ever types of tricks they may utilize.
This has already been beaten to death, but I'll say it anyway.
a) Military is only a tiny percentage of the budget when you consider that that we are NOT a socialist country (yet). Meaning if we did take on socialized medicine, higher education, etc. then the taxes would necessarily increase and the military percentage
b) The money that we spend on military almost exclusively goes back into American hands (whilst many other nato nations have to contract out to other nato nations such as Poland having a large percentage of it's entire budget going to the purchasing of foreign NATO planes). Thus the military is big business and usually fosters employment.
c) Our military budget is only big because of the cold war. The entire 90's were a down-sizing period. We're a naturally isolationist state (for better or worse). The main reason we do so many "peace keeping" missions is because we have a surplus military with little or nothing to do and a political leadership that determins the cost of war for us is cheap due to sunk costs (plus the viability for battle-hardening). Sadly, we're spreading ourselves too thinly and pissing everybody off in the process. This isn't helped by our current cowboy administration who pretends like they're invincible and that no other country's "feelings" are worth a damn.
As as for tax rates.. It has been said that Americans are the most "honest" tax payers in the world, so if you take that rumor into account, then we probably do pay comparibly higher taxes than other nations. Obviously this doesn't apply to "honest" citizens, but as an aggregate this does count.
Slashdot Mirror
If they can maintain a 1:1 performance on existing code, then the extra price-tag should be warrented for people that actually need it.
If the costs are no more than 1.5:1 for the 64bit CPU, then they can gut the IA-32 only line. Future die-shrinks should bring down costs of the 64bit chip and make it close to 1.1:1 cost, at which point they'll be posed to make 64bit their main line. Marketing wise, they'll be able to say, "we have a chip comparible in price/performance to intel, BUT we have 64bits".
The danger, of course is that any speed-ups applied to the 64bit archetecture (for 32bit code) can be applied to their athlon line (or it's eventual 32bit replacement). Thus they can always produce faster/cheaper IA32 chips, hurting the AMD64 demand.
The only reason to market AMD-64 to the desktop is to try and hurt Intel (priming the market and setting a standard). But they would have to forego lucrative "server" prices.
It's actually funny because it's generally rare to be able to apply newer chips into older motherboards. In other words, this is a solution that has an artifical problem. That PC chips were so cheap and easily available that it was worth maintaing a rapidly outdating sub-system (which comprised an increasily important role in overall performance).
I know several non-PC vendors attempted to use CPU-cards, where the CPU/cache subsystem were a mini-motherboard, to minimize the upgrade pains. Interfaces between sub-boards were standardized, etc. Problem is of course:
1) added MB cost
2) radical architecture changes require upgrading the interface (Pentium II and up can be thought of having this problem since CPU+Cache+MMU were on one mini-board)
3) anti-marketing... Why buy a chip when you can buy a whole new computer (and resell your old computer intact).
I actually gave up on the x86 CPU upgrade path a while ago when I started on Linux.. Instead of spending good money on bad, I kept the aging PC on the network but removed it's monitor.. Strapped Linux on it and played with Beowolf. I've owned several dozen machines in my day (propably purchased a couple dozen more CPU's for my early days of CPU-upgrades).
Except that it's highly non-portable. The types of optimizations you can make with segmented memory spaces generally affect the entire operating environment. Theoretically, the OS could provide hooks that return not only an address, but the associated segment register. Now you have entire applications requiring segmentation. (Doesn't win-16 code do this? I only ask because I see segment-register resource counts in Norton System Info.)
I believe the IA-32 segmentation representation utilized addition to produce physical addresses. While this allowed arbritrarily sized and placed segments, it introduces a performance hit. It's been a long time, and correct me if I'm wrong, but I believe the VM paging system is stuck with addition instead of "and"ing as well. So there is a very real cost that came about from this delayed transition to flat memory spaces.
IP-ID is not for reconstruction but identification.. It's like a database RowID.. Sequentiality has no bearing except that auto-incrementation is easier to implement than true random number uniqueness. mssql and other databases, however actually do perform random number generation (checking against a database of previously used values).
In any case, IP-ID is good for ICMP requests (such as ping)... If someone drops your ping's echo-request packet, they can be nice and send a response saying I've gone and dropped your IP package uniquely named IP-ID. Many servers don't do this anymore for fear of the "ping-of-death" DOS attack.
As for your question of how to reassemble packets. This is only really relavent for non TCP protocols (ICMP, UDP, etc). TCP has it's own sequence counter and can easily facilitate mini-sized IP packets. UDP can theoretically send 64k messages however and relies apon the underlying IP to fragment.
My memory is a bit fuzzy, but there are additional IP fields that specify which fragment out of n-total a given packet is. The idea is that routers/gateways can reconstruct/break-apart the packet arbitrarily at each leg of the internet's journey. I believe that each fragment still carries the originating IP-ID.. But it is the IP-ID + fragment-number + timestamp that uniquely identifies a packet (since a server will eventually reuse old IP-ID's).
The problem with filtering is that most people have "good mail" go into their inbox and spam mail go into a spam folder.. This is ludicrous for several reasons.. What if you're on a mailing list that sends dozens of emails a day (perl-6, for example). What if you miss who the sender is and you make assumptions based on the subject line or header, only to draw incorrect conclusions.
The answers my friends are to folderize your emails. As best as possible EACH sender should have their own inbox. This removes the subtle issue of temporary mistaken identity (since you have to physically navigate to their folder). More-over, for spam addresses, they can easily be flagged as spam and go into a sub-category that easily gets closed off.
One issue, however is the priority of filters (ordering for evolution/kmail/etc). If you use 3'rd party spam filters, make sure that the very first action is to weed in recognized addresses. This avoids your particular problem. The only caveat is that some viruses are sent from trusted senders. Thus any virus checking (which here is only currently relavent to windows [l]users) should go to a sender.virus folder, or rename the subject line to VIRUS: $orig_subject.
Unfortunately company-wide virus checkers are difficult to reconfigure in this manner. At our site, we do the subject-rename and let the email pass through. Thus our workstations can easily handle such rules however we choose.
While I don't know about the Rubidium-80 reference, the difference is quite possibly the poly-exclusion principle. I have only casual knowledge in the topic, but it seems to me that interaction-based quantum entanglement would involve time-space-state cohabitation which fermions generally can not do (by definition).
Thus, if you can transport the state (spin, charge, etc) of an atom and impose it on another atom, you have effectly transported the atom, even though no mass travelled the distance.
Except that this isn't necessarily useful with respect to either faster-than-light travel or moving massive bodies about, since quantum entanglement requires initial contact as far as I understand.
It is probably useful for secure communication since you initiate a beam in the middle going in opposite directions, then manipuate the photons at one end, and read the photons at the other (I realize it isn't this simple). Here it isn't the bandwidth, but the state of the photons that are of interest.
Actually, that would be RMS.. Linus was simply a saintly advocate of the GPL relgion.
> cmd.exe?
.NET API of an application. Essentially administrators could be programmers at a lesser level (only having to learn the very public APIs (such as print, convert image attributes, etc) similar to OLE features). It's much more powerful since application developers don't have to think about what to export; if they write inter-application API's, it'll be available to the sys-admin.
I think the point was that windows doesn't facilitate command-line friendly applications (You're more encouraged to do GUI and only GUI).
Therefore, no shell can compensate for the lack of glue that application developers provide.
It does seem that MS is on the right track though. Assuming that administrators are willing to learn the ins and outs of VB.NET, they'll have access to the entire
Course the same thing already exists with java, but there just isn't enough java in OS's tool-boxes to really be considered here.
Ethical problems don't exist... It's a made up thing, a simple case of "I believe this, so I will force it upon you".
Only if you're a firm believer in chaos. While I'm not a sociologist, it seems to me that laws are enforcements of local social norms.
Obviously we don't put you in jail for masterbating (well, in private at least), yet that can be argued as mass genocide (an entire potential sub-race). On the other end, we tend to get upset when a 1 year old is tortured to death.
The gray line in between can only be drawn by such social norms. The only thing I know for sure is that ANY such lines should not be uniform. Different regions of the world are composed of different biases, and thus would give you different stopping points of acceptability. So long as we don't have a single common law for such limits, then I think the world will be fine.
The grand unification (aka federalism / globalization) stands to get in the way.
Part of the obvious problem here is the bias to-wards a soul. Some believe the Soul is introduced at conception, while others at first sign of activity. Most of us probably treat a baby like a chicken-McNugget; We eat the McNugget so long as we don't recognize the dead cow in its shape/smell/color. We're happily ignorant of the horrible life-style of said cow. Taking the life of something that doesn't "feel" human is easier to avoid guilt/shame. Of course, still others don't believe in a soul, so the problem becomes a practical one.
In any case, the "Ethics" do in fact exist, but in a subjective form. This doesn't diminish their importance in decision making.
Personally I believe all living cells share the same life-force; whatever that might consist of (be-it an elaborate bio-electronic circuit, or some netherworld spirituality). It's merely human bias that places human life over that of a mouse (a bias that's justifiable, since other higher sentient forms will also sympathize/defend their own). A human cell in a mouse's body is still a human cell, and still alive. Human DNA in another being's body isn't much different. The trick; I think, is that when you have a sentient being, then you remove it's sentience, then it will experience death (loss of sentience). To what-ever degree we can determine that threshold, I would feel would draws the line of morality. Otherwise, the individual cells that make up a pre-child are nothing more than worker cells; such as blood-cells, hair follicles, etc. While their life is precious, they have little value atop that of saving the lives of thousands though brute-force disease research.
There are those that feel that stem-cell research and aborted fetus based research will encourage abortion-for-profit. Thus, even if the subjective matter of non-potential-life tissue isn't strong enough, they'll claim secondary immoralities. My response is that regulation can easily be applied to restrict such actions. Social norms will undoubtedly prevent mass-human synthesis (at least for the time being). Even if the future holds some perverse sub-human programs, I'm sure that our child-like incompetence of the intricate workings biology will facilitate normalizing natural disasters (such as the synthesis of new diseases, or any other sci-fi outcome).
I agree with what you're saying.. And I was in no way putting down literatures et. al. I was merely saying that I was never inspired in those fields, and so I never excelled. Thereby being evidence of the importance inspiration (given my attributes in other inspired fields).
More-over, I never said to home school. I merely said that teaching has a definite limit; you can bring a horse to water, but you can't make him drink. I guess I neglected to add the value of a mentor; someone that can answer frustrating questions.
The best way I can summarize my intentions is to say to simply be a role model for your children. If they want what you are and have, then they may have inspiration to follow in your tracks and take on the passions that you have. If your passions are TV, video games and socializing, then you can't get too upset when your children don't do well in math. If you obsess over software programming and don't portray a practical benifit to them, they may simply ignore you; writing you off as an eccentric.
As a youth, my father could work wonders with electronics and cars. We had propaganda about how great Einstein and Benjamin Franklin were and what they were able to do in their lives. There were the stories of rags-to-riches by various people throughout American history. These things provided me inspiration; and being an introvert, I wasn't AS distracted by the social aspects (though there was a lot of TV). My point is that it's a delicate balance. Tiny things can influence monumentally for better or worse.
I guess one active thing a parent can do is to keep ready track of who their children's mentors/idols are. They're generally open about them/ willing to discuss them. Course, when they fall in love with a rock-star or other such hard-to-reconsile idol, I don't know what one can do. But as you said, you have to let them be their own person.
If you really want to edumicate your kids, do it yourself.
Actually, while I don't know what kind of studies there are for it, the best way of "having a child learn", is when they learn for themselves.
Teachers necessarily can't know how to talk to a brick wall effectively, yet that's exactly what we bring generation after generation up as.
In com sci classes at least, the people that require tutors, generally don't do well; requiring spoon-feeding and "can't you just do my project for me". Those that get deep into the material on their own can often ace at least introductory course-work (com sci can definitely be a mind bender, especially when dealing with number theory).
The key is motivation. If you are unmotivated when you meet a person, you may remember little about them. Names are most famous: Many people (myself included) are already pessimistic about being able to remember hair/eye-color, names, etc. so I don't bother paying much attention when I meet someone.. Sure enough, it's a self fullfilling prophesy.
People going to computer or math classes with such pessimism have little chance of succeeding (regardless of their background deficiencies).
In my life, I've found that having desirable projects that happen to require learning a lot about a given topic affords an ample amount of motivation. In science, I crave sci-fi concepts; I want to understand them so I could possibly invent something new. With computers, I develop overwhelmingly complex goals (on the MRPG scale). Thus virtually every aspect of science, math and computer skills have been on my "I need to know" list.
Conversely, I haven't found such motivators for history, art, music, literature, so I only give those subjects a necessarily passing glance. (Though at some point I developed an appreciation for the story-telling nature of history).
While being totally non scientific (effective sample pool of 1), I still see such trends, and believe that inspiring your children in the single best way to teach them.
The trick is of course, how to inspire. And how do you avoid making a project obviously contrived to the point of frustration.
I'm afraid not. E-Mail allows me to send a message, or respond when I want to. Much better flexibility than IM.
Actually, when I used ICQ, I admired it's treatment of messages as mini-emails. If you were offline when you got a message, it would be available for you when you logged back in.
Therefore, it's perfect for sending offline important messenges that need greater priority than spam-neighbored emails (which people classicaly think to check periodically instead of continuously).
Effectively, ICQ was equivalent to an email client with a heirarchy of per-sender mail-boxes, where only the most activly recieved are up front (such as a spline tree). If you could set the "you've-got-mail" equivalent-tone to only activate when a top tier (say 10 senders) give you new mail, then you'd effectively have the same thing, though for high-volumen, it wouldn't be as efficient (due to TCP session per message-group, and header over-head).
To not repeat myself, here's another
posting
Summary is that there's evidence of a holographic aspect to neural cognition, and thus raw TFlops are hard pressed to simulate it's computational complexity. Thus the brain is still far ahead of computers by a couple orders of magnitude.
Also, I would split them "vertically", not "horizontally": multiple competitors that each produce the OS and applications, not a single company for each product.
:)
I'm sure this topic has been beaten to death, but I'm not quite sure how this would work (haven't thought about this in what must have been years).
Either:
A) give each fork a complete licence to the existing code, thereby forking all the software
B) mix and match segments, effectively doing both vertical and horizontal splits. (One doing MS Hardware and some but not all office products, but both doing Office/Windows (the main point of contention))
If A, then both companies would simply duke it out until one won (namely dirty incompatibility tricks, having even less incentive to share APIs). Note that winning=loss of profitability for one company.
Note if B, then you don't avoid the problems of A, but you tend back towards verticle non-competing companies.
I'm just not invisioning how horizontal could work.. With AT&T, you at least could regionalize the market and then regulate the leasing of regional lines to non-regional companies. But we're talking about a software commodity (at least at first).
Maybe it's just time for me to go to sleep.
You're talking about the substitution effect. The previous guy was talking about Supply and Demand.
.NET) which will of course require upgrading.
There is a certain demand for PC's. Some people can NOT afford a $2000 PC, no matter what the configuration of costs is. Many people can afford a $150 console gaming system, however. Likewise the demand for consoles is higher than for PCs. To some people, a MAC or Linux are not substitutes. Thus, they may simply elect to not make a purchase (this includes deciding to not purchase an ADDITIONAL PC for say their children).
Thus, you have a demand curve that's directly related to the price. Higher price means fewer purchased PCs. While there is a substitution effect, it is completely practical to say it is only a small factor (given the realistic monopoly of MS on our everyday computing needs).
Thus MS, being a veritable monopoly on a virtually zero marginal cost product (e.g. software) can greatly determine the price.
However, If they charged $1k for their OS, there would be fewer purchasers. Revenue = quantity_sold * unit_cost. At some point revenue will peek for a given price. The propensity for consumers to pay higher prices without losing too many sales is the elasticity curve. You can be absolutely sure that MS is keeping VERY close tabs on the elasticity of their product, and charging accordingly.
Further, they are working VERY hard to artificially enhance the elasticity curve. By regularly making OS releases, there is an irrational desire to "have the latest and greatest". Irrational only in the sence that experience should teach them that MS hasn't provided many good "new releases".. They've often been buggy and filled with "more of the same". Still some are attracted to the glitter of new graphical UI's, and they justifiably acquire some utility; though it is doubtful that that utility equals the cost of an upgrade.
MS goes one further by obsoleting existing software lines (due to non-backward compatibility). New games require new Direct X engines, which aren't provided on successively older OS versions. Office is necessarly non backwardly compatible. MS seeds new markets which will generate interest in new software paradigms (such as
All this equates to maintaing a heafty elasticity and demand curve. This more than compensates for the substitution effect as it exists today.
There is one good thing to come from all of this.. As they price themselves closer and closer to their desired elastically justified point (held back only by sticker shock psychlogy, and fear that the government will intervene), they lower the "barrier-to-entry" costs that were fundamental to their anti-trust-trial. It becomes ever cheaper relatively-speaking for a company to choose Linux. Moreover, it becomes more viable for a 3'rd party to charge money for a newly developed UI for Linux. Moreover, companies (such as Win4Lin, and others) are already doing it.
The only obsticles are compatibility.. But such a commercial company is now legally required to have access to such compatibility information from MS (as far as I understand the most recent rulings).
Unfortunately, MS knows this, so their cost increases will probably continue to stagnate; being just above what is acceptible, but not above that critical barrier to entry.
The article's paper is jiberish in the sence that they are merely demonstrating what we already know.. MS is not constrained in terms of pricing by competition. I was also upset to read what MS "could" have charged, since this neglects the billions spent in R&D.
The problem is that Windows currently takes the role of stanrdard oil or AT&T.. A necessary comodity that is being charged well beyond the reach of needy people. Walmart is selling $200 PCs ($300 if you want windows). That's 50% of the cost for a barely useful machine (web browsing and that's it). To write a term paper, they'd have to shell out the equivalent of 200% of the cost of the machine. The paper is obviously dated in this respect. What we really need is a paper on the effects of harm to the public due to excessive charges.
The OS itself is only worth maybe $25-$45 dollars, but MS doesn't just sell an OS. They provide lots of "innovation" for additional fixed-cost. The monopoly argument must be tied to the harm of the population (as with AT&T / standard Oil) for our corporate friendly Republican congress to give this issue a second glance.
Monopoly is the ugly boil of capitalism that nobody wants to admit is real. ...
According to contemporary economic theory, profit only occurs where a monopoly exists.
While I've minored in economics (and don't pretend to be an expert), I have a few beefs.
First, due to diseconomies of scale, it's entirely possible to regionalize a product. Franchises are somewhat related. Independent ownerships (and pricing) with royaltiest paid. Technically, McDonalds wouldn't be a monopoly, so long as it was regionalized.
Moreover, I've always hated the concept of zero profit. There is a tremendous false impression that zero profit means no money is made. The simple fact is that executives earn a salary, and that salary is counted as part of the fixed and or variable costs (variable if they charge hourly (which counts if they are compensated via bonuses)).
In a perfectly competative market, theoretically any possible reduction of variable costs lowers the market price. Thus, companies with lower executive pay-offs would affect such a change (given the greater proportion of their salary v.s. a workers salary). In reality, however, there are few truely "perfectly competative" markets. This is polluted from things like brand-name recognition, or similarly risk-aversion costs (due to not purchasing from known name brands), or locality/availability. Computer fairs, for example would be a good candidate for perfectly competative markets, but several dealers are shady, and those vendors set up next to the front door are highly likely to charge more for the same product.
The fact that many such computer companies are family owned (possibly even with family connections overseas for lowered costs), the employee costs are directly tied to firm profits. Yet, obviously if there was zero increase in price from OEM purchase to customer sale, the companies would all go out of business. Yet, somehow it's still worth the while of hundreds of companies to visit these computer fairs.
The reasoning is that everyone must generate at least a penny of revenue in excess of hardened marginal costs (meaning raw materials). That penny necessarily goes to the variable cost of labor. But that labor is directly tied to "profit" of the owner. It is then the goal of marketing / business strategy to command greater than 1 penny "profits".
The point here was merely debunk the notion of zero profit and perfect competition. Again, technically all labor expenses should be summed into the profit category since they are variable and expendable in direct proportion to the level of competition.
Beyond that, I believe that the concept of Average cost and short/medium run plays heavily into the practicalities of profit maximization. Theoretically in the long run, Average cost equals Marginal cost, and therefore price = MC due to competition. From this, only Monopolies can maintain a profit. However, no product/consumer is static. Virtually every product has evoked change in supply or demand to a degree that warrants both newer R&D costs (contributing to higher Average costs) and to short-term niche markets which allow for higher consumer elasticities.
I am thinking of the entry to maret as a diffusion rate with a natural impedence (resistive) force. The rate of change of a market is directly proportional to the ability to command profits, since there will be more opportunities to command a niche with little competition.
Further, given the set of markets that are changing (arguably a majority), and that AC = MC only after an infinite time, then for this set, AC != MC necessarily. Thus the dynamics of such transient markets are too complex to generalize. For some firms, costs will be lower for market entry, and thus AC for them will be lower than for competition. However, that market will eventually saturate or be obsoleted. At that point, transition can occur where the process begins again.
Such a system is highly susceptible to marketing (information scarcity / irrational purchasing), assymetric advantages (family connections, illicit collusion, etc), and simple luck (being at the right place at the right time at the right price with the right product).
My point in all the above is merely that the rate of change of a market is sufficient to allow for non monopolistic / non-perfectly-competative markets that could very well have been either. More-over, the food industry / computer industry necessarily falls into this category of rapid change (new "better" products replacing older ones).
I'm sure a lot of research has gone into this area, and I freely admit that I'm out of my leage. I'm simply [trying to] debunk what seems to be the elementary/classic view. You "can" make a "profit" from your hard work; you simply have to get past the idea that your dividends from your IPO'd company are the be-all-end all (completely ignoring the lunacy of capital gains off IPO'd shares, which is nothing more than gambling). (And yes I realize that economic profit covers more than dividends, but it is merely a different scale).
My main motivation for the above rant is the notion of fixed wages or worse yet, salaries w/in a corporation. Company effibility (to outside investors / analyzers) ignores the economic welfare due to employeeing potentially thousands of people with very high salaries/wages. If, instead, there were low salaries and instead a heavy bonus structure directly tied to contributed value to the company (i.e. productivity), then expenses would be more variable, and the need disperse profits would deminish; bonuses could be proportional to profits, and annual reported losses would be significantly reduced. The system rewards productivity retroactively instead of proactively. Mistakes / setbacks are more quickly regulated in terms of firm performance-analysis and future decision making. Most notably, there are fewer sunk costs.
I'm ranting at 2am, so I appologize. The more general argument, however is the viability of smaller, regional firms with little or no stock-based control / revenue streaming. The goals are a reduction in what I consider the adverse wants/effects of corporate economies of scale (namely corporate immorality/ neglect/devaluing of consumers) which is a natural extension of the unregulatable masses of businesses in a market based economy.
They charge you $1.25 for seventeen cents of syrup and some essentially free carbonated water.
While it's true that soda has a high elasticity (and hense viability for high profit margins), this is only a tiny fraction of the cost of a restaurant. McDonalds isn't exactly doing well these days, and I'm seeing more and more restaurants charge higher and higher prices for their soda (fountain or not). The issue is that while the marginal costs for McDonalds are very very low (cheap labor, mass produced fries, burgers, soda), the Average costs are very high (economic-rent, expansion costs, pyramidal corporate heirarchy costs, dividend payouts, etc). McDonalds is only really cash positive during their rush hours. But so long as they're open during slow periods of the day, they're losing money.
Like any rational being, they are maximizing the revenue based on elasticity curves.. And the fact that restaurants as a whole charge fabulously for soda reduces the substitution effect and thereby augments the elasticity.
Note I said reduces, and not eliminates.
Though I have grown completely tired of McDonalds, I'm still a poor college student, so what I do is buy bulk Soda from Grocery stores while they're on sale (I'm too poor to even join a Sam's club). Then I go to McDonalds and purchase from their Dollar menu, and bring the food home and drink my bulk soda at $0.25 / can. The net effect is that I take advantage of their speedy service, decent volume/taste and most imorptantly, regionality (one located every so many miles).
Unfortunately McDonalds does not reap the benifits from people like me. I generally acquire a great deal of value from my purchase, and most likely press up against their marginal costs, to say nothing of their average costs. I can see the effect as they slowly discontinue items from their dollar menu (such as their double cheese burger; arguably, their greatest value for some time).
I know, it's faulty logic on MS's part; I recently bought licenses to all my illegal MS software because the university was selling them cheap. Before, I couldn't afford office and windows XP and vis-studio.net, so I stole them. Then I paid about 50 bucks and got licenses.
MS knows what they're doing. While they lose SOME money to piracy, what they ARE doing is using scare tactics to ensure compliance from businesses, who command fabulous elasticities. A couple hundred dollars per employee is nothing to a business; especially considering the cost of Office furnature (even a computer these days is trivial; I should know, we run a small business, and have been plagued with such growth costs). The risk of being audited (these days) has a much higher associated cost (especially when factoring in the bad publicity to clients), and therefore it's physically cheaper to maintain compliance.
This goes for both Office and the OS, which is why it's harder and harder to even be able to afford MS Access (being bundled with ever more expensive packages). The problem to people like me is that I am made less valuable when consulting for friends / businesses by not having affordable access to the latest technology (while they (being part of a compliant, modern company) are generally limited to non-backward compatible versions).
I have absolutely no problem with MS attempting to copy-protect their own works (via online registration, etc.). This issue makes it exceedingly difficult to attain educational copies of Office / MS SQL, etc. Note that I'm not kidding about the word educational. I don't use MS for any personal activities, nor contracted activities.. However, being a software developer, I feel the need to compare/contrast and consult on MS technologies. Their "student editions" are crippled and bare little resemblance to their production products. Therefore, the only way to "learn" their material is to outright purchase full copies. This is of course unacceptible for such marginal use (e.g. it would be cheaper for me to purchase linux HW for my perspective clients than to purchase the latest MS SQL to lear how to help a friend setup such a machine).
However, I am actually one of MS's targeted platforms. In order to sell myself, presumably, I must be a MS developer (paying $1k / year to MS), and I must regularly certify myself and thereby take certification classes. Given that I feel that MS products are inferior in general (though there are certain superiorir attributes, relating to UIs), I do not wish to lock myself in. I choose UNIX development (Solaris/Linux/IBM, etc), but sadly this is not fully practial.
Thus, like you, I will sorely miss the ability to acquire evaluation copies.
While it is surely possible to "crack" most anything. The legal ramifications are becoming increasingly costly. Moreover, as I've repeatedly learned, cracked copies may contain maliscious code or may simply be crippled in unforseen ways. My girlfriend was unable to install Photoshop among other products on a cracked version of XP. Eventually I recommended purchasing an OEM copy and life was better (especially considering lowered tech support issues).
The end result of this disertation is that Nash/Adams is upheld here. We are right in secrely pirating (meaning we must be dignified in accepting defeat if we are caught), and MS is right in thwarting our piracy. A system where you can't cheat is fundamentally better than one where you shouldn't cheat.
Well put. Just a few additions.
A capitalistic society is necessarily a buyer-beware society. One of adverse selection, and price/quality volitility.
However, it is also a game of rationality. With the Adam Smith (society benifits from each acting according to his best interests), or better John Nashes addition of collective barganing, we can compensate for corporate abuses by:
1) Having a reliable system of regulatory law
2) Having an adaptable set of regulations
Part 2 is qualitative at best, but generally should work for major issues (supposedly MS should have been such a case).
Part 1 is the most important. In developing nations, it's less likely to exist. Mafia, corruption, beuracracy, and fear of hurting the economy by punishing opportunistic yet successful businesses all frustrate this goal. However, even after successfully enforcing laws, it is important to make sure that there is a very low probability that a corporation can "get away with" violations. Lacking such accountability, corporations can factor in such prosecution into the cost of risk. Prosecution is no more coercing to them than, say, the risk of losing a couple shipments of supplies/product due to bad weather.
Prosecution should be nothing less than detremental to a business, and there should be no [risk of] leanency(sp?), for fear that the process can be treated as insurable loss. Arthur Anderson is a wonderful example of such a practice. It'll be another decade before an accounting company would risk such data-munging.
This is the only way that a mostly free capitalistic society can prosper.
Personally I feel that elements such as software patents are egregious, but these fall into the qualitiative elements of regulation. So long as item 2 is upheld, however, the long term should not be the worse for it all.
Well, and breaking up big monopolies is a good first step towards addressing that problem and restoring a free market.
While I'd love to see MS broken up; I highly doubt that it would solve much. It would only get deregulated, given enough time. MS and Intel are in collaboration (due to mutual advantage). Why would OS / Office companies be any different.
Office, for example, proportedly would better support Apple / Linux. BUT, Linux support would cost them money initialy (and would require nightmarish tech-support), and would open up the Linux platform. Doing so leaves them vulnerable to competition, such as from Star Office. Further, I believe many proposals had share holders owning stock in all sub-divisions. They would obviously further the collusion efforts in any ways possible.
Remember, we're living in the same as as Enron, with accounting and inter-corporate cohabitation that puts even MS to shame. Any circumvention is possible.
Lastly, no government is going to break up MS any time soon; at least not until we're out of a ression. MS is a large contributor to our international exports; thereby lessoning the trade-deficit.
I remember seeing one billion dollars thrown around several times for the advertising budget.
I might be off, but I remember a $1B advertising campaing related some a MS product recently; might have been WinXP. The analysis of it was that this statement was pure marketing.. They included in that money expenditures of OEMs, and other creative aspects. That particular campaign was more like $200 Million physical MS dollars.
Given the ruthless rational marketing gaming they're playing, I wouldn't trust any numbers that come from MS. Remember that they have to justify their [low] expenses, not only to the government/public, but to their share holders; many of whom have been crying for years to have dividends paid out.
Actually if you're smart about it, you don't have to deal with spam if you're willing to risk losing mail.
Simply treat your inbox as a pseudo-junk-mail folder. Basically repeated spam gets filter rules to go to junk-mail/be deleted (I'm paranoid and never delete mail; disk space is cheap). Periodically scan the inbox for new contacts which should get their own personalized folder+filter, and scan for new bulk contact domains to add a junk-mail-filter.
I get between 25-200 messages a day, 90% junk-mail, and I never have a problem corresponding with close friends.. Some times a rarely-talked-with associate will have to wait a week before I scan through the inbox to find him as non-junk-mail, but I find that acceptable.
Go to any online computer vendor (Not sure about Dell), and notice the purchase of Win XP Home Edition at around $89 OEM. Vendors that perform bulk transactions may have special secret deals that cost them less, but such vendors are notorus for charging 100-1000% profit-margins on accessories (such as for 128Meg of RAM). Built a better custom machine for a friend than Dell was offering for about 2/3 the price, and that INCLUDED XP Home pre-installed. The only difference is the potentially extended warrenty offered by Dell (and of course, brand name recognition). I even give thumbs down to Dell and friends web site, since they don't give up front pricing on components (mainly package deals).
Anyway, point is, unless you are limiting your options (e.g. Windows-aligned dealers), then Windows isn't a sunk cost.
If SRAM were as cheap as DRAM we'd be using it for system memory and might not even need cache at all.
Not necessarily. There's an inherent slow-down associated with large address spaces. Not to mention the heat decipation. Heck, why else do we have 3 to 5 layers caching? The practical approach is to have successive layers of cheaper, larger and slower memory.
Since we already have 8 meg caches (in some high end machines), there's little value in doing away with multi-gig low-power, low-cost memories. Theoretically some apps will achieve noticable performance gains, but at enormous costs (today at least).
Furthermore, DRAM with internally managed refresh logic is functionally identical to SRAM (but non-deterministically slower). For something like video memory which regularly touches every byte of memory, the refresh logic would be unnecessary; thereby speeding up the memory. Further, DRAM is sufficiently performant enough to handle refreshes. 4MB * 80fps (for true color 1280x1024) = 320MBps. DDR can handle 2.1GBps alone. This doesn't even acknowledge the possibility of interleaving/banking/segmentation or what-ever types of tricks they may utilize.
This has already been beaten to death, but I'll say it anyway.
a) Military is only a tiny percentage of the budget when you consider that that we are NOT a socialist country (yet). Meaning if we did take on socialized medicine, higher education, etc. then the taxes would necessarily increase and the military percentage
b) The money that we spend on military almost exclusively goes back into American hands (whilst many other nato nations have to contract out to other nato nations such as Poland having a large percentage of it's entire budget going to the purchasing of foreign NATO planes). Thus the military is big business and usually fosters employment.
c) Our military budget is only big because of the cold war. The entire 90's were a down-sizing period. We're a naturally isolationist state (for better or worse). The main reason we do so many "peace keeping" missions is because we have a surplus military with little or nothing to do and a political leadership that determins the cost of war for us is cheap due to sunk costs (plus the viability for battle-hardening). Sadly, we're spreading ourselves too thinly and pissing everybody off in the process. This isn't helped by our current cowboy administration who pretends like they're invincible and that no other country's "feelings" are worth a damn.
As as for tax rates.. It has been said that Americans are the most "honest" tax payers in the world, so if you take that rumor into account, then we probably do pay comparibly higher taxes than other nations. Obviously this doesn't apply to "honest" citizens, but as an aggregate this does count.