Detecting CPUs and cores

[Roedy Green]

It makes sense to charge according to the VALUE a customer gets from
your product.

It seems illogical to me that I pay the same for a program I hardly
ever use, and one I use continuously. You need some sort of scheme to
share the cost pie of providing the service. So long as the seller
has not wangled a monopoly position, he should be free to cook up any
scheme he pleases. The market place will punish the irritating
schemes..

 

[Twisted]

"profits equal to that which would be generated by the normal rate of return,"

And what, precisely, is this "normal" rate of return? If it's more
than the bare minimum needed to keep the company afloat, then
something's inefficient and someone's probably getting ripped off,
including the consumers. If your total costs in providing me a foobar
would be $1, then charging much more than $1.05 or so for that foobar
seems excessive; you get to pocket a nickel, and consumers like me get
to have the foobar for roughly what it costs to provide one, and not
much more. Charge much more, and a) your competition can undercut your
price and still turn a profit, and b) consumers like me have to pay
much more for a foobar than what a foobar costs to provide, which is
unfair. People who can actually afford the marginal cost of a foobar
may nonetheless be denied access, without justification since they can
bear the cost of providing them one by hypothesis. Unfortunately, lack
of sufficient competition makes this all too common and results in the
poor having much less access to things than they should, given they
could pay the marginal costs; it artificially inflates the cost of
living which does nobody any good in the long run.

Of course, if you're like most of the posters here you make your money
selling non-commodity software at fat margins and hate the idea that
all useful tools should be available as commodities, even though such
would be vastly enabling to people and, in the longer term, vastly
increase the rate at which the economy generates wealth by lowering
costs and barriers-to-entry. Of course, lowering barriers to entry
also generates more competition, which as businessmen you'll all
really hate too...

 

[Twisted]

The money paid to those programmers, and associated costs (office space,
workstations, unlimited free drip coffee, administration expenses,
benefits...)

Workstations are reusable, you know. Or do you mean the electricity
costs from running them, and suchlike?

Software tends to be a high risk business, more like venture capital
than stable government bonds.

An odd claim, given that the costs can clearly be made virtually zero,
as evidenced by the sheer quantity of software that is generated by
hobbyists, or as part of the free software movement, etc.; the barrier
to entry and minimal costs given reasonable efficiency are quite low.
I wonder if the high costs you cite are peculiar to the development
environment of commercial software houses, where the coders have
clueless managers looking over their shoulders constantly (and the
managers themselves demand ludicrously high salaries given they don't
do anything that resembles anything that I'd call "work"), and they
are made to go to an expensive, inefficiently heated and cooled,
inefficiently lit office building to work instead of permitted to do
it on their home computer, and ... etc. etc.; open source is often
vastly more efficient it seems, where the "managers" giving
requirements information and feedback to the developers are the users
themselves, and the developers can use their existing floor space and
equipment, etc. so there's no big expensive tower full of cubicles; no
manager salaries; with no attempts at claiming proprietary "IP" and
other such BS there's no lawyers to pay; there's just the coding and
the distribution of copies. It's clearly vastly more efficient. It's
no wonder Linux and OpenOffice are starting to seriously threaten
Microsoft, if Microsoft's development and delivery are inherently
inferior and inefficient!

 

[Twisted]

It seems illogical to me that I pay the same for a program I hardly
ever use, and one I use continuously.

It seems illogical to me that I pay different amounts for things that
cost the same amount to provide to me; and it seems like a ripoff any
time someone insists that I pay vastly more for something than it
costs to provide me with it. Not to mention unfair.

You need some sort of scheme to share the cost pie of providing the service.

You're comparing apples and oranges here. You're discussing something
like electricity, where the costs of providing it scale up with their
usage; or net connectivity, or similarly. I.e. a *service*.

I'm talking about a computer program; a piece of executable code that
performs some useful function when run. Whoever wrote it and shipped
me a copy has a single large cost when they make the first ever copy,
and a single small cost when they make and ship my copy. They don't
have any ongoing costs proportional to my subsequent usage, assuming I
don't bother them for support or anything.

I should be able to buy or otherwise get this copy and then use it for
as long as it will function, the same as if I bought a car. Of course,
software can be kept functioning potentially indefinitely, modulo
keeping backups or the original install media, avoiding incompatible
updates to the operating system or whatever, continuing to have
hardware it runs on, etc.

So long as the seller
has not wangled a monopoly position, he should be free to cook up any
scheme he pleases. The market place will punish the irritating
schemes..

Including any scheme that charges a lot more than cost, and certainly
any that charges based on usage when the "provider"'s costs don't grow
proportionately to your usage but rather remain constant.

 

[Patricia Shanahan]

Workstations are reusable, you know. Or do you mean the electricity
costs from running them, and suchlike?

No, I mean the cost of owning or leasing the workstation for the
duration of the software development. If I were to try to sell, now, a
computer that cost $1,000 a couple of years ago, I would get a lot
less than $1,000 for it. If I had invested the money two years ago in US
Government bonds scheduled to mature today, I would get more than $1,000
back.

That difference combines two forms of cost, the depreciation on the
workstation and the opportunity cost of the capital used to purchase it.

If I leased the workstation the company I leased it from would have made
similar calculations, and those costs would have been included in their
charges.

Patricia

 

[nebulous99]

No, I mean the cost of owning or leasing the workstation for the
duration of the software development. If I were to try to sell, now, a
computer that cost $1,000 a couple of years ago, I would get a lot
less than $1,000 for it. If I had invested the money two years ago in US
Government bonds scheduled to mature today, I would get more than $1,000
back.

[snip rest]

A strange attitude. Might be the result of renting your hardware
instead of buying it. You could just buy it and then develop as much
software on it as you pleased. A common off the shelf home computer
that's a few years old (but no more) is perfectly adequate to run
Eclipse and whatever other development tools you need. Of course,
anything you develop that has steeper system requirements will require
a different testing machine.

Another place where FOSS development seems to get superior efficiency.
It gets developed with hardware that's generally "free", in that it's
a sunk cost purchased for other reasons, and tested on same, and even
more, on the end-users' machines. FOSS development basically doesn't
involve any hardware expenses that are purely for the purpose of the
FOSS development.

 

[Patricia Shanahan]

No, I mean the cost of owning or leasing the workstation for the
duration of the software development. If I were to try to sell, now, a
computer that cost $1,000 a couple of years ago, I would get a lot
less than $1,000 for it. If I had invested the money two years ago in US
Government bonds scheduled to mature today, I would get more than $1,000
back.

[snip rest]

A strange attitude. Might be the result of renting your hardware
instead of buying it. You could just buy it and then develop as much
software on it as you pleased. A common off the shelf home computer
that's a few years old (but no more) is perfectly adequate to run
Eclipse and whatever other development tools you need. Of course,
anything you develop that has steeper system requirements will require
a different testing machine.

That just changes the amount of money per workstation, not the fact that
computers depreciate with time or that invested money increases in value
with time.

Patricia

 

[nebulous99]

That just changes the amount of money per workstation, not the fact that
computers depreciate with time or that invested money increases in value
with time.

It makes the ongoing costs of the development workstations only what's
needed to keep them in operating condition, since you need never
replace them.

I notice you don't address my remarks about how FOSS seems to develop
far more efficiently, with those front-loaded costs vastly lower per
unit of functionality than commercial software development.

 

[Patricia Shanahan]

It makes the ongoing costs of the development workstations only what's
needed to keep them in operating condition, since you need never
replace them.

But you still paid for them up front, losing the money you paid for them.

I notice you don't address my remarks about how FOSS seems to develop
far more efficiently, with those front-loaded costs vastly lower per
unit of functionality than commercial software development.

If there happens to be free software that meets all your needs and has
acceptable licensing terms you should definitely use it.

The difficult case is when there is no free software that does the job,
and too few people, relative to the size of the program, feel like
developing it for fun. That is when software starts costing money, as
you are dealing with programmers who insist on being paid to do that
task, and provided with office space, equipment, etc.

Patricia

 

[=?ISO-8859-1?Q?Arne_Vajh=F8j?=]

If by that you mean "a thin margin on top of parts, labour, shipping,
and handling" then fine.

No it is a profit equivalent what else profit you could have gotten.

If we go from macro economics to business, then consider it the
interest you could get from putting your money in a savings account
or invest in government bonds. If you business gives less profit than
those, then you will (at least if you are profit maximizing) put
the money in them.

Arne

 

[Guest]

Note that expenses in this context includes "normal profit".
[...]

No, I think that she means "profits equal to that which would be
generated by the normal rate of return," which makes more sense since
you should put your money elsewhere if it can't get you the normal rate
of return.

P.S. I am sorry if I got the gender wrong. I had assumed that Patricia
had made the earlier referent post, and in my speed, I forgot to verify
that fact. I humbly apologize.

No hard feelings.

Arne

 

[Bent C Dalager]

If there happens to be free software that meets all your needs and has
acceptable licensing terms you should definitely use it.

The difficult case is when there is no free software that does the job,
and too few people, relative to the size of the program, feel like
developing it for fun.

The vast majority of such requirements are met by in-house development
projects, for which licensing issues are a moot point. The costs are
covered by revenues from the company's main business.

Cheers
Bent D

 

[Twisted]

No it is a profit equivalent what else profit you could have gotten.

If we go from macro economics to business, then consider it the
interest you could get from putting your money in a savings account
or invest in government bonds. If you business gives less profit than
those, then you will (at least if you are profit maximizing) put
the money in them.

Sounds like a more competitive economy would automatically have lower
interest rates then, as interest rates must themselves be tied to the
rate of return on investments. So the substantial amount of
uncompetitive monopoly rents that can be extracted right now raises
the rate of return on some things and makes every other business seek
to raise prices to keep up a similar rate of return, and ends up
raising interest rates and, presumably, causes inflation in the longer
term.

This suggests that ending all monopoly rents would cause some initial
upheavals, but eventually a low-inflation and low-interest but rapidly
innovating and wealth-generating economy.

Who stands to lose in such a situation, that they'd oppose (with overt
or covert machinations) such a development? Well, it looks like the
whole banking industry, as well as the actual monopolies...

 

[Joshua Cranmer]

It makes the ongoing costs of the development workstations only what's
needed to keep them in operating condition, since you need never
replace them.

Under your guidelines, you would be saying that the ACME Company should
be using its vintage UNIVAC from the 1950's that is easily outclassed by
my TI-89. Even in more modern cases, replacement of workstations is
probably needed.

Take my laptop: it is (was at the time of this scenario) about 5 years
old. I wanted to upgrade to Windows XP (required for certain pieces of
software). Guess what? It has 64 MB of RAM, gotta go upgrade that! And
there's more: half of its hard drive is unusable due to corrupt sectors.
(It has a damaged keyboard as well--the shift key came off). By that
time it becomes cheaper just to get a new computer.

If you continue to insist that workstations need never be replaced, then
tell me where to find a 1 GB stick of PC-133 MHz DDRAM (so that more
than 1.5GB of RAM can be supplied to a computer).

 

[Twisted]

It makes the ongoing costs of the development workstations only what's
needed to keep them in operating condition, since you need never
replace them.

Under your guidelines...[snip]

URRGH! Will you stop ARGUING and just LISTEN? The people deserve
access to tools and commodities at prices not much elevated above the
marginal cost of providing them. End of story. Finito. You cannot
argue against that without being pro-monopolist and plutocratic, or
even communist, rather than true-capitalist.

 

[Roedy Green]

I'm talking about a computer program; a piece of executable code that
performs some useful function when run. Whoever wrote it and shipped
me a copy has a single large cost when they make the first ever copy,
and a single small cost when they make and ship my copy. They don't
have any ongoing costs proportional to my subsequent usage, assuming I
don't bother them for support or anything.

The argument is should I base the price on MY cost or the VALUE the
customer gets from my work? Capitalist theory says you should raise
your price to the point you start making less money because of the
lower sales.

The end user is not all that interested in your costs. Whether you
were blew a bundle on start up costs, or did it for a song, and what
your incremental and support costs are, he consider YOUR problem.

Basically he is interested in the value HE extracts from your program.
If customer A is able to extract more value out of a program than
customer B, A will be willing to pay more than customer B. SO it
seems logical if you can find a way to charge A more that B, you are
going to earn more money, than if you used a flat cost where some
customers did not buy because it was too expensive and others coasted
happy to pay considerably less than the program was worth to them.

This partly why I suggest using rental and an service as the basic
mechanism for delivering computation. The more hours per month, the
more seats, the more the customer should pay. They are getting more
value.

Any scheme is arbitrary. The schemes that makes the creator the most
money and keeps the customers happiest and gets the programs most used
are the ones I would prefer to use.

 

[Twisted]

The argument is should I base the price on MY cost or the VALUE the
customer gets from my work? Capitalist theory says you should raise
your price to the point you start making less money because of the
lower sales.

You also omit to mention that capitalist theory says that in a
competitive market you won't be able to raise it very much above your
marginal costs.

You also omit to mention that capitalist theory says that an
uncompetitive market is a Bad Thing(tm).

SO it
seems logical if you can find a way to charge A more that B, you are
going to earn more money

You neglect to note that doing so is underhanded and dishonest.
Advocating discriminatory pricing? Shame on you! How consumer-hostile.
And again a properly competitive market won't let you get away with
shit like that. If you offer the same thing to B at a lower price B
can turn right around and sell it to A for a penny more, turn a
profit, and you quit being able to sell to A at a high price as he
buys from B instead of from you. Of course, you have to raise the
price for B somewhat to make up for this, but instead of charging $50
to B and $150 to A you charge $100 to everyone, more-or-less exactly
as it should be modulo the other effects that should apply to your
pricing.

The only way you can stop this is by doing something truly dirty, like
using legal bludgeons or gratuitous technical "features" to make the
thing you sell to B non-transferrable to A.

But capitalist theory says that owning one's private property,
including being able to resell it or otherwise dispose of it as one
sees fit, is a good thing.

So what you're contemplating is bad. Naughty, naughty Roedy!

This partly why I suggest using rental and an service as the basic
mechanism for delivering computation. The more hours per month, the
more seats, the more the customer should pay. They are getting more
value.

It will never fly unless you force it somehow, because customers will
always pay less in the long run by buying a copy than by renting a
copy, and will pay even less by using a FOSS program instead of paying
anyone anything.

So you would have to outlaw FOSS and ownership of copies of software
in order to get your way. So much for being a capitalist; a capitalist
must accept the existence of such competition, including competition
that undercuts your price, or sells people an eternally functional
copy of some software that does the same thing you'd charge them
monthly to do.

Too bad, so sad.

You cannot be allowed to have your way or it is the end of the 21st
century "knowledge economy"; people like you, given the chance, would
strangle it in the cradle to ensure their own profits at everyone
else's expense. Make all my software much buggier and more expensive
will you?! Over my dead body! You can have my personally-owned and
eternally functional software when you pry it from my cold, dead
fingers! Plentiful, cheap, commodity software is a feature, not a bug!

Any scheme is arbitrary.

Until the market applies corrections to prices. Your get-rich-quick
scheme hinges on it not doing so, and therefore on some pretty nasty
coercion. May you fail spectacularly.

The schemes that makes the creator the most
money and keeps the customers happiest and gets the programs most used
are the ones I would prefer to use.

Then why do you suggest the schemes that make the creator the same
amount of money as before**, make the customers saddest*, and gets the
programs least used** then?

* Because you steal their software and rent it back to them, and they
must pay lots more for the same benefits, and must give up some of
their software because the costs have shot up so dramatically, plus
you hold their data hostage, and they have lost more of their property
rights in their things.
** Because people have a certain total budget for spending on
software. The more expensive it is on average, the less it gets sold.
So in fact software makers don't make any more money than before. They
make that same amount, only they do so while providing less software.
OK maybe their costs go down so they do show a small profit relative
to before, but it won't actually work to get rich quick; sorry to
burst your bubble there. And software will be much less used. In fact,
if usage is metered people will severely ration their usage.

Your suggestion makes the whole world poorer. The whole world! Even
YOU!
Customers are poorer because they must pay more for the same stuff,
and since they only have so much money they actually have to pay the
same amount for less stuff. So they have less stuff.
Worse, stuff they could just pay for and then own they now are forced
to pay over and over again for. Would you like it if your household
furniture could not, through some legal artifice, be yours, and had to
be rented by law? You couldn't buy chairs or tables or a desk; you had
to rent them. After a while the rental adds up to much more than the
price of buying one and still you have to keep paying or they
repossess your chairs and tables and desk. I don't think you'd like
that would you?
Plus there's the intrusive and gratuitous enforcement scheme. Given
it's designed to prevent the software from working some of the times,
this DRM will make the software especially prone to the worst category
of bugs, showstopper bugs. Also there's an incentive to hold the
user's data hostage to force them to keep paying you for the privilege
of being able to access it; this you'll spin as an "anti-piracy"
measure to stop people cracking the software for indefinite use, while
neglecting to point out how it also creates a lock-in that will let
you jack up prices until you bleed everyone dry. And even then there's
really only so much money to go around, and the end result is you end
up no richer and everyone else ends up poorer.
Oh yeah, and you also end up poorer. First, because of all the
software YOU use whose costs have shot through the roof, both directly
in price-tag terms and in terms of showstopper bugs introduced by all
that evil DRM. Second, because the whole WORLD ends up poorer, and the
sinking tide drops all boats, including yours.

I wish you bad luck with this scheme, I really do.

 

[Joshua Cranmer]

URRGH! Will you stop ARGUING and just LISTEN?

I am sure that many people would be happy to ask you do to that as well.

> The people deserve
access to tools and commodities at prices not much elevated above the
marginal cost of providing them. End of story. Finito. You cannot
argue against that without being pro-monopolist and plutocratic, or
even communist, rather than true-capitalist.

You can neither argue against that if you are pro-monopolist,
plutocratic, communist, or whatever. The term 'deserve' automatically
makes your argument an appeal to emotion and not a valid point because
it is a subjective and unquantifiable quality. Besides, the article
'the' is definite and therefore refers to a specified subset of people;
you have not given this subset.

I suppose, in retrospect, that one could define a utility metric that
measures how well-off people are and use that as a definition: "people
deserve X" would be equivalent to "the utility metric Z would be higher
if X were the case", but then the validity of utility metric Z itself
would be called into question. In any case, define your position in a
more verifiable way and then

In any case, a product selling for "little more" than the marginal cost
will not be found in the long term assuming that a given company will
take whatever measures it can to maximize profit, within bound of the
law. If it at any point in time it cannot make the normal rate of
return, the corporation will find that it can make more money by putting
its money elsewhere (like some safe T-bonds) and will withdraw from the
market. In the process, supply of said product will decrease, increasing
the equilibrium price. Aggregating the suppliers together means that the
price will equalize at the level needed to achieve normal rate of return.

 

[Bent C Dalager]

In any case, a product selling for "little more" than the marginal cost
will not be found in the long term assuming that a given company will
take whatever measures it can to maximize profit, within bound of the
law. If it at any point in time it cannot make the normal rate of
return, the corporation will find that it can make more money by putting
its money elsewhere (like some safe T-bonds) and will withdraw from the
market. In the process, supply of said product will decrease, increasing
the equilibrium price. Aggregating the suppliers together means that the
price will equalize at the level needed to achieve normal rate of return.

In a truly competitive marketplace (as is generally considered
desirable in capitalistic theory), all products will see only a
miniscule profit on top of production/distribution cost. This is one
of the beauties of capitalism in that it provides useful, inexpensive
products compared to most - perhaps all - other economic models.

In such a market, however, it is difficult to make a huge profit and
become insanely wealthy. This is considered by many to be a problem.
There are some tried and true methods of circumventing this problem
however, including (but certainly not limited to):

* Becoming immensely huge so that even one cent per item turns into
millions on the bottom line. (E.g. Wal-mart)

* Sinking lots of investment money to get into an industry with high
barriers to entry. (E.g. cutting edge microchip manufaturing)

* Operating in a market with artificial government imposed
monopolies. (E.g. entertainment, software and general invention)

Twisted seems to be having a go at the latter and saying that these
industries should be forced to compete in a free market rather than
ride the government gravy train.

Cheers
Bent D

 

[Bent C Dalager]

You neglect to note that doing so is underhanded and dishonest.
Advocating discriminatory pricing? Shame on you! How consumer-hostile.

This is quite common and some of its current implementations seem to
work reasonably well. It's generally called market segmentation and
attempts to exploit various mechanisms that encourage wealthy people
to spend more on the product than what the less wealthy are prepared
to do.

A few examples would include:

1. Price-drops over time. You see this as a matter of course in the
computer games market. Games will come out at a high initial price
($50 perhaps) and slowly drop down over time, eventually ending up in
the bargain bin ($10 to $20 perhaps). The wealthy people will buy it
the day it comes out (preorder it even) at full price while the more
stingy will pay less, later.

2. Mail-in rebates and coupons in general. These are annoying and take
a fair bit of time and attention from the customer. They are therefore
not appealing to the wealthy customer who has better things to do with
his time (and whose hourly wages may even make it a financially
superior option to work an extra 15-30 minutes rather than take that
same time handling the rebates/coupons) whileas the less wealthy will
often find them a highly valuable supplement to their domestic budget.

While these may be seen as evil mechanisms, it should be considered
that they also contribute to making products cheaper than they
otherwise would have been to poorer people and personally I don't
consider that a bad thing.

And again a properly competitive market won't let you get away with
shit like that. If you offer the same thing to B at a lower price B
can turn right around and sell it to A for a penny more, turn a
profit, and you quit being able to sell to A at a high price as he
buys from B instead of from you.

The problem is, the well-off people don't care about those savings
anyway, so the market for B is really rather limited. In fact, A may
find it more valuable being able to brag to his friends about how he
buys all of his games (say) at full price anyway. Humans have always
liked to flaunt their wealth when they have it. Admitting that you get
your stuff cheaper by having your neighbour collect coupons or
whatever may be socially embarrassing.

Cheers,
Bent D