(
Me An advantage you might get out of using C++ is that tight loops may compile down into much faster code than you would get with Perl.
Most of the CGI programs that I've written don't even have loops.
... Which is precisely why this isn't all that huge an advantage.
And which is a reason why it might be preferable to use a language like Lisp that can compile to machine code, thereby providing the best of both worlds:
Efficiency, by having the parts that need to be fast be compiled, and
Reasonable treatment of dynamic objects like strings
An advantage you might get out of using C++ is that tight loops may compile down into much faster code than you would get with Perl.
Unfortunately, you lose some abilities:
The ability to change a script "on the fly" whilst debugging, and have the change automagically deployed. With C++, you have to make and then go through whatever installation process is required to deploy the change.
With web applications, what you're largely manipulating is text, which means that having the language oriented to that is extremely valuable. Furthermore, since there are powerful, well-optimized operators built-in to these languages, the interpreter disadvantage is significantly diminished.
Some are rather particular to formats used in mainframe applications, but some sure look like they'd represent common practice, things any programmer worth his salary should be able to recreate...
Some have suggested that RHAT should dump money onto Debian.
This begs the question of whether the Debian folk would find this to be of any value... For the most part, Debian doesn't consume money. If they suddenly had $8M to throw around, this would enter "monied politics" that could have pretty shattering effects.
And, in any case, why would anyone consider it appropriate for RHAT to need to sponsor a project that is quite directly competitive?
It would be quite appropriate for RHAT to bounce some money to the FSF.
The FSF actually has offices, budget, and some experience in hiring programming staff. Which makes it a more sensible idea, in some ways, to throw "ludicrous amounts of money" at the FSF than it would have been for Debian.
But even still, $8M is on the order of 30 times the recent amounts of annual funding of the FSF.
Again, dumping $8M on the FSF would be not unlike pouring a truckload of chocolate syrup onto an ice cream sundae. Interesting to see, but likely rather messy and even wasteful.
Similar could be said for big "dumps" onto XFree86, LaTeX3, Linux International, PostgreSQL and other nonprofit orgs.
I'd rather see RHAT hire a bunch of people and build software that is released under free licenses, with CVS archives to allow external contributions as well. The costs will be as "tasty" a tax writeoff as anything else; the benefits include that if they regard the results as valuable, perhaps others will find them similarly valuable.
Strange, how that sounds a whole lot like RHAD Labs...
you can't deny that the jump is predictable and real.
Can you provide a reference to a statistically validated study to support this claim? All splits on the NYSE are reported; it should be simple enough to do a study of the price increases resulting for all the splits over the last ten years. I'd expect to see such results in some place like the Journal of Finance; it would doubtless be a feather in the cap of someone wishing to overturn the Efficient Market Hypothesis.
However, those that actually study such things (as opposed to those that are out to sell you their Technical Analysis Newsletter) find things like the following:
"Elaborate tests of the correlation of successive prices, runs, and filter rules find some weak relationships, but they are not sufficient to generate trading profits after taking account of transactions costs."
- Graham and Dodd's Security Analysis
It seems entirely more likely that if stock prices continue to rise after a split, this results not from the split itself, but rather for whatever reasons there were for the stock to rise in price before, perhaps because the enterprise is continuing to reap unexpectedly high profits.
In some respects, I'm just as glad that I don't have a vote, as the choice between the options of "I claim I didn't inhale!", "I will not answer whether or not I used coke," and "I'll bodyslam my honorable opponents!" doesn't admit a clearly reasonable choice.
Based on that and on local "fun and games," I'm not particularly surprised that the voter turnout for Dallas' last election was, for a city of nearly a million people, what I used to consider poor turnouts for elections of school board trustees back in Ottawa...
The problem isn't merely of ignorance; apathy can arise when it's not clear that the vote cast will be of any useful value...
Hank the Angry Drunken Dwarf might be our opportunity to get the "people don't care enough about security" situation out in the open.
Independent of whether the system is secure enough or not, I certainly agree that there are merits both to:
Making voting easier
Since many elections are showing off lower and lower levels of participation, and
Keeping the effort required high enough
... so that people take their own vote seriously...
Highly automated voting apparatii are more strongly associated with Neilson ratings, and thus with determining whether Oprah or Jerry Springer are more popular...
Of course, Hank the Angry Drunken Dwarf could be a better candidate than some that have come along...
While Portfolio Theory provided quite unsurprising results, the way that Black/Scholes provides differential equations that usefully analyse what was thought of as statistical matters was pretty amazing, and has helped employ a surprising number of theoretical physicists in finance.
I would put index funds at the top of my list of "investments to consider" simply based on Harry Markowitz's 1952 Journal of Finance paper, Portfolio Selection. He didn't anticipate index funds yet at that time, but they're a pretty ideal representation of his construction of "efficient frontiers" and "optimal portfolios." (And I had his paper quite specifically in mind when I used the words "efficient" and "portfolio" in the same sentence...)
For the "compleat idiot," an excellent book on investing is A Random Walk Down Wall Street; it provides a reasonably friendly walk through modern finance theory, and happens to rank index funds fairly highly for use by "nonprofessional investors."
I like the idea of starting with a portfolio that's largely index funds, and gradually adding to that a reasonably diverse stock portfolio, as that allows avoiding the administration fees that mutual funds (of whatever variety) charge; that of course requires taking Buffet's position of "buying stock in order to hold it indefinitely."
As far as I'm concerned (and I have a sizable essay to this argument ), the more organizations trying to do cool stuff the better.
It would be nice to see some funds get "thrown" at organizations like the FSF, SPI, XFree86, and so forth. Unfortunately, "throwing" money at them doesn't forcibly accomplish useful things. It is a joke that "Mathematicians are nature's way of turning coffee into theorems."
Frankly, I'd rather see RHAT take the simpler tack of hiring more programmers to write software that gets GPLed. Note that that happens to be just as much of a "tax break" as a contribution to a charitable organization.
The merit of having a separate organization may be the goodwill that results from having what may appear to be a bit more independent organization.
Alternatively, they may (and this is a bit of a reach, I'll admit) be founding an organization so that they can encourage others to contribute to it.
Consider: I'm not in the market for a "fully supported $80 RH Linux boxed set." But I might be game to help contribute to development efforts. If RHAT spins some/all of the development effort out to a separate organization, that may encourage people like me to contribute to that.
This could be a step towards transforming RHAT, the commercial organization, into the service organization that represents where IPO documentation has indicated that their revenues were expected to come from in the long term.
They roll development into a "charitable" organization, which:
Makes them look good,
Separates it from the profit-oriented activities,
Perhaps pulls in some of its own revenue stream.
Perhaps I'm being dramatically overoptimistic here; it's still possible that this be quite a good thing.
... And so any given single step is not going to have any vast effect on things.
But the only way there will be a "death of MSFT" is via the Death of a Million Paper Cuts that involve not a single "killing" blow, but rather a whole array of tiny, relatively independent injuries that add up.
Other comments have suggested that a more logical step would be to push for Linux pre-installed on (say) Seagate hard drives.
Put all of these things together:
Some copies getting deployed via bundling with motherboards
Some copies getting deployed via bundling with disk drives
I once got paid to do finance theory research work, working on code that compared Black/Scholes calculations with the results of computations coming from a similar calculation using discrete steps. I presently have about $10K invested in index funds. It would be pretty silly for me to imply that index funds are only for stupid people, and that's certainly not what I said.
Index funds most certainly are decent investments for diversifying away risk; those that invested in index have often done better over the last couple of years than those investing in other forms of mutual funds.
The point was that those that don't have the basic knowledge ( e.g. - to know that Value = Quantity * Price ) to invest in stocks with a faint bit of intelligence should put their money into investments that don't require that attention.
We've got people who haven't the rationality to evalate what they're doing deciding that they can become "Day Traders" at E-Trade and AmeriTrade. I don't oppose people buying some stock; that's liable to be educational and diversify ownership. But naive new investors certainly shouldn't jump into the shark-tank of daytrading. That's a really dumb move.
The nature of buybacks is somewhat questionable; it most certainly amounts to a manipulation of securities pricing.
On the other hand, the complaint about stock splits seems to me to be devoid of real validity.
The critical equation in dealing with portfolio valuation is:
Value = Price * Quantity
The average split results in multiplying quantity by some factor; while we might quibble over the "strong form" of the Efficient Market Hypothesis, markets certainly appear to have been reasonably efficient at recognizing that when the number of shares gets multiplied that price needs to be correspondingly divided.
Remember that the number of shares that are outstanding are an artificial construct; the number of shares issued is arbitrary.
Perhaps it would be better to use the phrase on into the indefinite future rather than forever, as that admits that there is uncertainty in the future.
Two very valid (and relevant) points that come out of Buffet's strategy include that:
If you have an investment that you approve of enough that you consider it worth buying, then it should be worth holding on to.
Churning stock in and out of a portfolio on a daily basis makes the broker rich from brokerage commissions. That doesn't make the stockholder rich...
The "split" issue appears to me to be a pretty irrelevant matter in this; the contention that splits result in people overvaluing stock seems to me to be irrational.
I don't particularly care what the stock price is, or how many shares I have; what I care about is the multiple, namely How much my holdings are worth.
As far as I'm concerned, a split should only affect the value of my holdings if there are a lot of people that don't understand basic arithmetic ( e.g. - multiplication) and who are incapable of coping with the simple equation:
'Creative accounting' is used by most major corporations, Billy G didn't invent it.
Quite true. The essential claim in the report is that the widespread use of stock options to compensate employees turns things into something of a "pyramid scheme."
Other articles claiming the same have come along from more credible sources. Of course, if MSFT is "guilty" of "pyramiding," there are also a whole lot of other companies that are also guilty, particularly with the ludicrously highly valued activities that fall out of corporate mergers.
If "everyone else is guilty too," this undercuts this article's claims somewhat, as it implies that if MSFT is overvalued, then many other securities are similarly overvalued.
Which doesn't make Microsoft any righter, but does suggest that they're just a convenient "whipping boy" for someone's political concerns.
My point would be that it's not enough merely to have "a UNIX environment;" that makes it a natural next step to have "some form of distribution" in the manner of a Debian/EROS that would parallel the existent Debian/Linux and Debian/Hurd systems.
That buys you not merely a somewhat familiar environment, but also a goodly chunk of the sizable toolset that people are accustomed to using on Linux.
The key to making EROS useful to people running Linux would be to build a "GNU System" atop EROS, parallelling building a "GNU System" atop the Linux kernel.
Anonymous Coward wrote:
If we stuck with this attitude from the beginning, we'd still be using JCL and COBOL for everything. It's a new system, how about new tools?
And if we want to accomplish anything with the new system any time soon, what are we to do?
UNIX has been remarkably persistent over the year as a useful environment for prototyping.
If there's nothing in common between EROS and Linux, and, in particular, if there are practically no tools available for EROS, there's going to be little merit to making a leap over to EROS.
The point here is that there are two particularly relevant outcomes:
EROS starts from scratch, having no tools with which users would already have any familiarity.
That makes for a challenging transition, and, for a would-be newcomer, they may say
This EROS system isn't even as functional as Linux. Why should I consider using it?
EROS provides some familiar UNIX tools, thus gaining a rich development environment, which makes it a more attractive transition.
Since EROS is different, it is obviously appropriate to add new tools in support of that. That is not synonymous with the contention that
We ought to throw away absolutely everything that we've found useful up until now.
I forgot to mention that using these "tax havens" to hide income is not, generally speaking, particularly legal if you're a resident of (say) Canada, the US, and other "less dodgy" nations.
If you're a resident of Canada, for instance, Revenue Canada considers your income for tax purposes to include all world income, and that includes whatever you might be trying to hide by flowing it through a tax haven.
Thus, if you don't leave Canada, probably taking on residency in one of the tax havens, the avoidance of tax represents tax evasion which is definitely against the law, and possibly a criminal act (as opposed to being "merely civil").
Becoming an International Criminal may not be on the "top ten list" of things to do this week...
The "offshore account" isn't terribly helpful to those of us that are employed; the money that comes to us has ample causes to be disclosed to organizations like the Internal Revenue Service (US), Inland Revenue Service (UK), or Revenue Canada (guess?).
Offshore accounts are useful in keeping transfers of funds hidden once that money is sitting offshore.
This would be useful for:
Making payments for questionable activities such as sales/purchases of illegal substances
Making payments for "consulting services."
Note that transferring funds to and fro between Transparent Jurisdictions like Canada or the US and Opaque Jurisdictions like Lichtenstein or Switzerland or... is:
Significantly inconvenient.
In order for the important transfers to remain secret, you'd need to have a whole pile of accounts, some to be used for the "Opaque Transactions," and others to be used for "Transfers To/From Canada."
One account is most definitely not enough, and the action of shifting money is also not convenient.
You don't use one of the "Hidden Accounts" to pay off your MasterCard bill, as that would expose its existence to domestic authorities.
Ditto for any other situations where you need to have money in Canada to spend on things.
Note that interest rates on deposits in these foreign places are not likely to be terribly high.
The tax havens are "capital rich," which means that based on supply+demand, interest rates will be low to nonexistent.
The net effect of this is that you don't actually make money off of having money in the tax haven; all you do is to avoid being taxed at home.
The net result of all of this is that there's probably little point to opening "tax haven" accounts unless you've got at least a few million dollars kicking around, or unless someone who would find it worthwhile to move money to a tax haven owes you payment for significant services.
The people that would thus care about this might be:
The key to making EROS useful to people running Linux would be to build a "GNU System" atop EROS, parallelling building a "GNU System" atop the Linux kernel.
(Note that I usually call "systems based on the Linux kernel" by the moniker Linux; the use of the RMS term happens to be usefully descriptive here; I'm not trying to do any politically-motivated Newsspeak here.)
I would tend to think that the Debian folks would be the most prepared to create an overall system atop EROS, as they have both
A set of automated tools for constructing and (to some extent) validating sets of packages, and
Some experience trying to fit Debian to a non-Linux kernel, namely Hurd
The major alternative that, based on the deployment of predecessor systems like KeyKOS, is likely to take place quite a bit, is that EROS might instead be largely used to construct "somewhat embedded systems" rather than the general purpose system that comes from installing the typical Linux distribution.
This might include:
Building a really secure little web server package
Building a really secure little file server package
None of the Network Computers thus far have been priced under about $800 USD.
Much of this has been because everyone was planning to build StrongARM-based systems where they knew the CPU cost only $25, failing to realize that the only way to get costs down was to have mass production of StrongARM motherboards and other components.
Net result: They hadn't yet generated quantities of product, so price wasn't down to an economical level to allow it to be cheap enough for anyone to even consider.
The sort of model that they need to follow is similar to that of the Nintendo 64/Sony PlayStation game systems; those units are getting sold these days for around $100-$150, and probably are sold at around cost. Several years ago, there was an April Fools Article on Linux on Nintendo 64; I've had a more serious assessment of this for a couple years now.
Down to details...
Way back when, everyone thought that they should be using "cheaper" StrongARM (or perhaps MIPS or PPC) chips that were greatly cheaper than the Intel stuff. The fact that you're left custom-building motherboards was the "kiss of death" to cheapness.
Now that prices of IA-32 chips have fallen through the floor, an Intel Celeron or AMD K6 may be economical enough.
The big deal is to have a compact IA-32 motherboard with integrated video, perhaps sound, and integrated Ethernet, along with some FlashRAM in lieu of a hard drive.
If there's a Taiwanese vendor selling such motherboards for $50 in quantity, add in $40 for CPUs, $15 for a plastic case, $40 for a stick of RAM, and $5 for power supply, and you've got a $150 internal cost.
That only leaves $50 for the costs of pushing the box through retail channels, which seems low. Of course, as with Nintendo 64 and Sony Playstation, the real money comes in selling software, and it will certainly be in Ellison's interests to have both service and software offerings for these boxes so as to extract more than $200 from the average user of them...
I rather hope that there will be some form of these lectures available in non-"Webcast" form, as that is a uniquely ephermal form.
Doubtless Knuth will be able to turn it into an engaging book; a followup of sorts to 3:16 would be a very nice thing to see; I'm sure I'll acquire it once available...
Although I still have to say that I'd put higher priority on at least a preprint Vol. 5/6 of TAOCP. That would provide at least draft form for all of the series, what with The Stanford Graphbase at least providing the flavor of Vol. 4.
Slashdot Mirror
And which is a reason why it might be preferable to use a language like Lisp that can compile to machine code, thereby providing the best of both worlds:
Unfortunately, you lose some abilities:
With web applications, what you're largely manipulating is text, which means that having the language oriented to that is extremely valuable. Furthermore, since there are powerful, well-optimized operators built-in to these languages, the interpreter disadvantage is significantly diminished.
Some are rather particular to formats used in mainframe applications, but some sure look like they'd represent common practice, things any programmer worth his salary should be able to recreate...
- US5758336: Date format and date conversion procedure using a packed binary format
- US5668989: Two-digit hybrid radix year numbers for year 2000 and beyond
- US5630118: System and method for modifying and operating a computer system to perform date operations on date fields spanning centuries
- US5761668: Method and apparatus for converting computer software and databases for the year 2000
- US5644762: Method and apparatus for recording and reading date data having coexisting formats
- US5737735: Method and apparatus for recording and reading date data having coexisting formats
- US5600836: System and method for processing date-dependent information which spans one or two centuries
Linux and Y2K...This begs the question of whether the Debian folk would find this to be of any value... For the most part, Debian doesn't consume money. If they suddenly had $8M to throw around, this would enter "monied politics" that could have pretty shattering effects.
And, in any case, why would anyone consider it appropriate for RHAT to need to sponsor a project that is quite directly competitive?
The FSF actually has offices, budget, and some experience in hiring programming staff. Which makes it a more sensible idea, in some ways, to throw "ludicrous amounts of money" at the FSF than it would have been for Debian.
But even still, $8M is on the order of 30 times the recent amounts of annual funding of the FSF.
Again, dumping $8M on the FSF would be not unlike pouring a truckload of chocolate syrup onto an ice cream sundae. Interesting to see, but likely rather messy and even wasteful.
I'd rather see RHAT hire a bunch of people and build software that is released under free licenses, with CVS archives to allow external contributions as well. The costs will be as "tasty" a tax writeoff as anything else; the benefits include that if they regard the results as valuable, perhaps others will find them similarly valuable.
Strange, how that sounds a whole lot like RHAD Labs...
Can you provide a reference to a statistically validated study to support this claim? All splits on the NYSE are reported; it should be simple enough to do a study of the price increases resulting for all the splits over the last ten years. I'd expect to see such results in some place like the Journal of Finance; it would doubtless be a feather in the cap of someone wishing to overturn the Efficient Market Hypothesis.
If your claim were true, then even the weak form of the Efficient Market Hypothesis would be false.
However, those that actually study such things (as opposed to those that are out to sell you their Technical Analysis Newsletter) find things like the following:
- Graham and Dodd's Security AnalysisIt seems entirely more likely that if stock prices continue to rise after a split, this results not from the split itself, but rather for whatever reasons there were for the stock to rise in price before, perhaps because the enterprise is continuing to reap unexpectedly high profits.
In some respects, I'm just as glad that I don't have a vote, as the choice between the options of "I claim I didn't inhale!", "I will not answer whether or not I used coke," and "I'll bodyslam my honorable opponents!" doesn't admit a clearly reasonable choice.
Based on that and on local "fun and games," I'm not particularly surprised that the voter turnout for Dallas' last election was, for a city of nearly a million people, what I used to consider poor turnouts for elections of school board trustees back in Ottawa...
The problem isn't merely of ignorance; apathy can arise when it's not clear that the vote cast will be of any useful value...
Hank the Angry Drunken Dwarf might be our opportunity to get the "people don't care enough about security" situation out in the open.
Independent of whether the system is secure enough or not, I certainly agree that there are merits both to:
- Making voting easier
- Keeping the effort required high enough
Highly automated voting apparatii are more strongly associated with Neilson ratings, and thus with determining whether Oprah or Jerry Springer are more popular...Since many elections are showing off lower and lower levels of participation, and
... so that people take their own vote seriously ...
Of course, Hank the Angry Drunken Dwarf could be a better candidate than some that have come along...
While Portfolio Theory provided quite unsurprising results, the way that Black/Scholes provides differential equations that usefully analyse what was thought of as statistical matters was pretty amazing, and has helped employ a surprising number of theoretical physicists in finance.
I would put index funds at the top of my list of "investments to consider" simply based on Harry Markowitz's 1952 Journal of Finance paper, Portfolio Selection. He didn't anticipate index funds yet at that time, but they're a pretty ideal representation of his construction of "efficient frontiers" and "optimal portfolios." (And I had his paper quite specifically in mind when I used the words "efficient" and "portfolio" in the same sentence...)
For the "compleat idiot," an excellent book on investing is A Random Walk Down Wall Street; it provides a reasonably friendly walk through modern finance theory, and happens to rank index funds fairly highly for use by "nonprofessional investors."
I like the idea of starting with a portfolio that's largely index funds, and gradually adding to that a reasonably diverse stock portfolio, as that allows avoiding the administration fees that mutual funds (of whatever variety) charge; that of course requires taking Buffet's position of "buying stock in order to hold it indefinitely."
As far as I'm concerned (and I have a sizable essay to this argument ), the more organizations trying to do cool stuff the better.
It would be nice to see some funds get "thrown" at organizations like the FSF, SPI, XFree86, and so forth. Unfortunately, "throwing" money at them doesn't forcibly accomplish useful things. It is a joke that "Mathematicians are nature's way of turning coffee into theorems."
Frankly, I'd rather see RHAT take the simpler tack of hiring more programmers to write software that gets GPLed. Note that that happens to be just as much of a "tax break" as a contribution to a charitable organization.
The merit of having a separate organization may be the goodwill that results from having what may appear to be a bit more independent organization.
Alternatively, they may (and this is a bit of a reach, I'll admit) be founding an organization so that they can encourage others to contribute to it.
Consider: I'm not in the market for a "fully supported $80 RH Linux boxed set." But I might be game to help contribute to development efforts. If RHAT spins some/all of the development effort out to a separate organization, that may encourage people like me to contribute to that.
This could be a step towards transforming RHAT, the commercial organization, into the service organization that represents where IPO documentation has indicated that their revenues were expected to come from in the long term.
They roll development into a "charitable" organization, which:
Perhaps I'm being dramatically overoptimistic here; it's still possible that this be quite a good thing.
But the only way there will be a "death of MSFT" is via the Death of a Million Paper Cuts that involve not a single "killing" blow, but rather a whole array of tiny, relatively independent injuries that add up.
Other comments have suggested that a more logical step would be to push for Linux pre-installed on (say) Seagate hard drives.
Put all of these things together:
- Some copies getting deployed via bundling with motherboards
- Some copies getting deployed via bundling with disk drives
- Some copies getting deployed via being preinstalled by one of the fifty-odd Linux VARS
And it starts to add up to "a few paper cuts."And the point is not to "beat Microsoft;" that would merely be a convenient sideeffect of doing useful things with Linux.
I once got paid to do finance theory research work, working on code that compared Black/Scholes calculations with the results of computations coming from a similar calculation using discrete steps. I presently have about $10K invested in index funds. It would be pretty silly for me to imply that index funds are only for stupid people, and that's certainly not what I said.
Index funds most certainly are decent investments for diversifying away risk; those that invested in index have often done better over the last couple of years than those investing in other forms of mutual funds.
The point was that those that don't have the basic knowledge ( e.g. - to know that Value = Quantity * Price ) to invest in stocks with a faint bit of intelligence should put their money into investments that don't require that attention.
We've got people who haven't the rationality to evalate what they're doing deciding that they can become "Day Traders" at E-Trade and AmeriTrade. I don't oppose people buying some stock; that's liable to be educational and diversify ownership. But naive new investors certainly shouldn't jump into the shark-tank of daytrading. That's a really dumb move.
People that choose to be this ignorant have no place investing in the stock market.
They should buy:
- Long term government bonds, or
- Index funds
both of which are more suited to the action of those that wish to remain blithering idiots.On the other hand, the complaint about stock splits seems to me to be devoid of real validity.
The critical equation in dealing with portfolio valuation is:
The average split results in multiplying quantity by some factor; while we might quibble over the "strong form" of the Efficient Market Hypothesis, markets certainly appear to have been reasonably efficient at recognizing that when the number of shares gets multiplied that price needs to be correspondingly divided.Remember that the number of shares that are outstanding are an artificial construct; the number of shares issued is arbitrary.
Two very valid (and relevant) points that come out of Buffet's strategy include that:
The "split" issue appears to me to be a pretty irrelevant matter in this; the contention that splits result in people overvaluing stock seems to me to be irrational.
I don't particularly care what the stock price is, or how many shares I have; what I care about is the multiple, namely How much my holdings are worth.
As far as I'm concerned, a split should only affect the value of my holdings if there are a lot of people that don't understand basic arithmetic ( e.g. - multiplication) and who are incapable of coping with the simple equation:
Other articles claiming the same have come along from more credible sources. Of course, if MSFT is "guilty" of "pyramiding," there are also a whole lot of other companies that are also guilty, particularly with the ludicrously highly valued activities that fall out of corporate mergers.
If "everyone else is guilty too," this undercuts this article's claims somewhat, as it implies that if MSFT is overvalued, then many other securities are similarly overvalued.
Which doesn't make Microsoft any righter, but does suggest that they're just a convenient "whipping boy" for someone's political concerns.
At this point, you've probably got:
There otta be a protocol...
That buys you not merely a somewhat familiar environment, but also a goodly chunk of the sizable toolset that people are accustomed to using on Linux.
And if we want to accomplish anything with the new system any time soon, what are we to do?
UNIX has been remarkably persistent over the year as a useful environment for prototyping.
If there's nothing in common between EROS and Linux, and, in particular, if there are practically no tools available for EROS, there's going to be little merit to making a leap over to EROS.
The point here is that there are two particularly relevant outcomes:
That makes for a challenging transition, and, for a would-be newcomer, they may say
Since EROS is different, it is obviously appropriate to add new tools in support of that. That is not synonymous with the contention that
Just a slight difference...
If you're a resident of Canada, for instance, Revenue Canada considers your income for tax purposes to include all world income, and that includes whatever you might be trying to hide by flowing it through a tax haven.
Thus, if you don't leave Canada, probably taking on residency in one of the tax havens, the avoidance of tax represents tax evasion which is definitely against the law, and possibly a criminal act (as opposed to being "merely civil").
Becoming an International Criminal may not be on the "top ten list" of things to do this week...
Offshore accounts are useful in keeping transfers of funds hidden once that money is sitting offshore.
This would be useful for:
- Making payments for questionable activities such as sales/purchases of illegal substances
- Making payments for "consulting services."
Note that transferring funds to and fro between Transparent Jurisdictions like Canada or the US and Opaque Jurisdictions like Lichtenstein or Switzerland or- Significantly inconvenient.
- You don't use one of the "Hidden Accounts" to pay off your MasterCard bill, as that would expose its existence to domestic authorities.
- Note that interest rates on deposits in these foreign places are not likely to be terribly high.
The net result of all of this is that there's probably little point to opening "tax haven" accounts unless you've got at least a few million dollars kicking around, or unless someone who would find it worthwhile to move money to a tax haven owes you payment for significant services.In order for the important transfers to remain secret, you'd need to have a whole pile of accounts, some to be used for the "Opaque Transactions," and others to be used for "Transfers To/From Canada."
One account is most definitely not enough, and the action of shifting money is also not convenient.
Ditto for any other situations where you need to have money in Canada to spend on things.
The tax havens are "capital rich," which means that based on supply+demand, interest rates will be low to nonexistent.
The net effect of this is that you don't actually make money off of having money in the tax haven; all you do is to avoid being taxed at home.
The people that would thus care about this might be:
The fork may have been necessary, and the eventual reintegration (or "reverse fork") that came from EGCS was also necessary.
But the initial fork displays that there were problems with GCC development that could not be reconciled at the time. And that was not a good thing.
The key to making EROS useful to people running Linux would be to build a "GNU System" atop EROS, parallelling building a "GNU System" atop the Linux kernel.
(Note that I usually call "systems based on the Linux kernel" by the moniker Linux; the use of the RMS term happens to be usefully descriptive here; I'm not trying to do any politically-motivated Newsspeak here.)
I would tend to think that the Debian folks would be the most prepared to create an overall system atop EROS, as they have both
The major alternative that, based on the deployment of predecessor systems like KeyKOS, is likely to take place quite a bit, is that EROS might instead be largely used to construct "somewhat embedded systems" rather than the general purpose system that comes from installing the typical Linux distribution.
This might include:
Which would parallel what Oracle has been working on with the "Raw Iron" Oracle 8i Appliance
I'd kind of like to see both approaches, as that is the most likely way for EROS to become more widely used.
Much of this has been because everyone was planning to build StrongARM-based systems where they knew the CPU cost only $25, failing to realize that the only way to get costs down was to have mass production of StrongARM motherboards and other components.
Net result: They hadn't yet generated quantities of product, so price wasn't down to an economical level to allow it to be cheap enough for anyone to even consider.
The sort of model that they need to follow is similar to that of the Nintendo 64/Sony PlayStation game systems; those units are getting sold these days for around $100-$150, and probably are sold at around cost. Several years ago, there was an April Fools Article on Linux on Nintendo 64; I've had a more serious assessment of this for a couple years now.
Down to details...
Way back when, everyone thought that they should be using "cheaper" StrongARM (or perhaps MIPS or PPC) chips that were greatly cheaper than the Intel stuff. The fact that you're left custom-building motherboards was the "kiss of death" to cheapness.
Now that prices of IA-32 chips have fallen through the floor, an Intel Celeron or AMD K6 may be economical enough.
The big deal is to have a compact IA-32 motherboard with integrated video, perhaps sound, and integrated Ethernet, along with some FlashRAM in lieu of a hard drive.
If there's a Taiwanese vendor selling such motherboards for $50 in quantity, add in $40 for CPUs, $15 for a plastic case, $40 for a stick of RAM, and $5 for power supply, and you've got a $150 internal cost.
That only leaves $50 for the costs of pushing the box through retail channels, which seems low. Of course, as with Nintendo 64 and Sony Playstation, the real money comes in selling software, and it will certainly be in Ellison's interests to have both service and software offerings for these boxes so as to extract more than $200 from the average user of them...
Doubtless Knuth will be able to turn it into an engaging book; a followup of sorts to 3:16 would be a very nice thing to see; I'm sure I'll acquire it once available...
Although I still have to say that I'd put higher priority on at least a preprint Vol. 5/6 of TAOCP. That would provide at least draft form for all of the series, what with The Stanford Graphbase at least providing the flavor of Vol. 4.