generate report in excel formate

[surajtheesun]

hi frns,
i just want how to generate report in excel formate,
actually i am taking data from database mysql and i ahve to generate
report in excel formate.
regards
suraj

 

[rossum]

hi frns,
i just want how to generate report in excel formate,
actually i am taking data from database mysql and i ahve to generate
report in excel formate.
regards
suraj

I am not sure, but I think that Excel can read comma separated
variable (CSV) files. You might want to do a quick test to see if it
can, because CSV files are very easy to create.

rossum

 

[TechBookReport]

I am not sure, but I think that Excel can read comma separated
variable (CSV) files. You might want to do a quick test to see if it
can, because CSV files are very easy to create.

rossum

Excel can read CSV files easily enough - but unless you can control the
locale CSV is not recommended. Different locales encode the decimal
point in numerics using different characters - the English-speaking
world generally uses a point '.' character, but some non-English
speaking locales use a ',' character. There also problems with thousands
markers.

From Java there are a number of options for writing XLS files directly.
The Apache POI project, the JExcelAPI library, Groovy and no doubt
plenty of others offer ways to do this.

HTH

Pan

 

[Cerveza Mas Fina]

You can use the library available from www.andykhan.com. It's pretty
easy to use too.

 

[hh]

For generate excel in java,jxcell is a good choice.
http://www.jxcell.net

 

[Twisted]

For generate excel in java,jxcell is a good choice.http://www.jxcell.net

ALERT! hh just attempted to stick a hand in your pocket. This jxcell
thing is not free. What is downloadable there is apparently crippled
in some way. DO NOT WASTE YOUR TIME. Continue searching for a free
alternative. Save your money. Don't subsidize subtle spammers.

 

[Roedy Green]

I am not sure, but I think that Excel can read comma separated
variable (CSV) files. You might want to do a quick test to see if it
can, because CSV files are very easy to create.

see http://mindprod.com/jgloss/csv.html

 

[Roedy Green]

You can use the library available from www.andykhan.com. It's pretty
easy to use too.

For others, see http://mindprod.com/jgloss/excel.html

 

[Roedy Green]

ALERT! hh just attempted to stick a hand in your pocket. This jxcell
thing is not free. What is downloadable there is apparently crippled
in some way. DO NOT WASTE YOUR TIME. Continue searching for a free
alternative. Save your money. Don't subsidize subtle spammers.

It is $300 US. The site has a PURCHASE button which makes it clear
the program is not free.

I don't know where you get his idea that people who develop software
don't need money for rent.

 

[Twisted]

It is $300 US.

Ludicrous. That's gotta be a whopping $300 more than it costs.
Whoever's selling it (hh?) is laughing all the way to the bank --
every purchase is pretty much pure profit margin.

The site has a PURCHASE button which makes it clear the program is not free.

However, the posting here by hh did not say anything at all to warn
the OP that their link would be worthless at best and a dangerous
trap* at worst for people without a spare $300 to blow on their
programming hobby.

I don't know where you get his idea that people who develop software
don't need money for rent.

I've never claimed anything of the sort.

* The OP might have not looked at the navigation buttons and just gone
straight for the download link. This is a link to a binary, not to
another Web page, so they'd just think they had the software and go to
use it. And then one day, bang -- it just tells them to pony up and
drops dead, leaving them in the lurch. They may have grown dependent
on it -- or worse, locked in to some proprietary data format it uses
for something -- and it's therefore too late for them to choose
something else, say something FOSS. Now they're forced to pay $300, or
suffer catastrophic data loss that might well be just as costly to
them if not worse.

Links to unfree stuff of any kind should be
a) Avoided. Not only are they less useful than links to free
alternatives, they are also potential traps (see above) and call one's
motives into question (are you getting a commission on every sale by
any chance, hh?)
b) If you really must, make sure to state up front in clear English
that people with spare money they're willing to blow need not apply.

This applies also to links to information behind paywalls, as well as
timebombed software. Stuff behind non-pay registerwalls too -- paying
in being deluged with spam or telling them where you live and your
favorite sexual position or whatever else their marketing dept. wants
is still paying, and it's jumping through hoops and wasting your time
to boot. (Gordon Beaton, who earlier and in another thread suggested
that I post something to a "newsgroup" that turned out to be a Web
forum requiring I sign up before I could post, take note. Even if I
didn't have to sign up, it would be a hassle having to regularly go to
an additional web site every few hours to look for replies instead of
just checking in my newsreader via the read-only nntp server I use for
reading news since GG has gotten so shoddy of late. I like having all
the replies in one place, my Thunderbird, and not scattered across a
zillion different Web pages, and I also like only having to hand out
private info to a) my ISP and b) Google and not c) a zillion
additional forum operators. It's bad enough that I have no news
posting ability that lets me mung my email. This gmail account gets a
thousand spams a week already as it is, which means anything that
mistakenly gets classified as spam gets deleted unread as I don't have
time to review 1000 messages a week for false positives.)

 

[Lew]

Ludicrous. That's gotta be a whopping $300 more than it costs.
Whoever's selling it (hh?) is laughing all the way to the bank --
every purchase is pretty much pure profit margin.

Impossible to say without seeing their books, since we do not know what it
costs the manufacturer. What's the overhead? What were the development
costs? For all we have evidence, they might be taking a loss.

 

[Roedy Green]

Ludicrous. That's gotta be a whopping $300 more than it costs.
Whoever's selling it (hh?) is laughing all the way to the bank,
every purchase is pretty much pure profit margin.

It is a very specific product, and he probably has sold no more than a
couple of copies. I have a similar problem with The Replicator which
has swallowed months of effort with only one sale.

 

[Twisted]

Impossible to say without seeing their books, since we do not know what it
costs the manufacturer. What's the overhead? What were the development
costs? For all we have evidence, they might be taking a loss.

If it's data, and they choose an efficient distribution mechanism for
this, then the marginal cost is pennies. They could sell copies for a
dime apiece and turn a profit in the long run. That they charge not
only more, but THREE THOUSAND TIMES MORE, means they lack viable
competition, and means a market failure of the "some asshole has a
monopoly and is being predatory with it" variety has happened. Again.

 

[Twisted]

It is a very specific product, and he probably has sold no more than a
couple of copies. I have a similar problem with The Replicator which
has swallowed months of effort with only one sale.

You'd have more to show for it if you charged less, I suspect. It
would surely be in more widespread use if you'd charged nothing. Of
course you'd have less money in that case -- albeit only one sale's
worth. But if you had charged something small, but nonzero, you might
very well have ended up with more money than you actually did using
your actual price point.

 

[Lew]

If it's data, and they choose an efficient distribution mechanism for
this, then the marginal cost is pennies.

You keep throwing around that term "marginal cost" as if you actually knew
what you're talking about. I was talking about all the costs, not just the
marginal cost of distribution. Even the most elementary materials on
accounting and economics point out that there are many costs, such as the ones
I explicitly mentioned (overhead, development cost). You keep harping on one
smallest category of cost as if it were the only one. Wrong.

Since you change the argument to base it on the least significant cost a
business faces, and pretend that there are no other costs to the business, you
present an utterly fallacious case.

You would excoriate the farmer for charging fair prices for produce because
their farmstand costs so little to set up. What about the whole growing
season of labor, when they're spending and spending and spending hoping their
crops will recoup the costs? Then you come along and tell them they're
monopolists because the "marginal cost" of a farmstand doesn't justify their
prices.

Your favorite pseudo-expert term "marginal cost" is useless. That is only the
/marginal/ cost, i.e., the cost /other than/ all the other costs you so
conveniently ignore. You'd have all software vendors out of business in no time.

Remember for future reference, "marginal cost of distribution" is not the
only, nor even the most significant cost of software for the manufacturer.

 

[Martin Gregorie]

If it's data, and they choose an efficient distribution mechanism for
this, then the marginal cost is pennies. They could sell copies for a
dime apiece and turn a profit in the long run. That they charge not
only more, but THREE THOUSAND TIMES MORE, means they lack viable
competition, and means a market failure of the "some asshole has a
monopoly and is being predatory with it" variety has happened. Again.

The thing you're forgetting it the cost of creating the software,
database, collated information or whatever. Unless you're a hobbyist
that cost needs to be recouped. There are a few ways:

1) Build a one-off bespoke item for one customer. He pays agreed
cost+profit and can then do what he likes with the item.

2) Charge the initial customer the full cost+profit and give
copies away to everybody else. That won't fly: nobody will pay
full price with that sales model, but that sounds like what
you're advocating.

3) Charge the initial customer the full cost+profit but sign an
agreement that they get a rebate every time you sell another
copy. I have seen this used but its rare because the initial
customer can be hard to find and the admin costs are
relatively high.

4) Spread cost+profit over a projected sales target, 'n' and charge
each customer (cost+profit)/n - the usual sales model BUT it
depends on preventing free copies being made and distributed.

5) Give the product free but charge for manuals, installation and
support. This also depends on nobody making free copies of
manuals, giving free support or (worse) third parties charging
for support. I've seen it used (e.g the Pegasus MUA) but its too
subject to petty pilferage to work well.

6) Err, I don't think there is a 6, at least not if you want to
earn a living from your work.

 

[nebulous99]

You keep throwing around that term "marginal cost" as if you actually knew
what you're talking about. I was talking about all the costs, not just the
marginal cost of distribution. Even the most elementary materials on
accounting and economics point out that there are many costs, such as the ones
I explicitly mentioned (overhead, development cost). You keep harping on one
smallest category of cost as if it were the only one. Wrong.

No. It is you who are wrong. You conveniently snipped the rest: "They
could sell copies for a
dime apiece and turn a profit in the long run."

The "other costs" you speak of are one-time costs. They spend these
once, then the marginal cost for each copy sold. So they're initially
in the hole $D, the development one-time costs. Subsequently, they get
$P-$M, price minus marginal cost, on each copy. Assuming $P-$M is
positive, the amount they're in the hole by shrinks linearly over time
and eventually goes negative. At that point they're showing a net
profit and from then on it's gravy train time.

It's elementary arithmetic. That you cannot grasp this shows that you
are not a fit opponent for this debate. Stop posting over and over
again arguing with me and forcing me to waste my time refuting your
drivel and unwanted, mathematically-bankrupt opinions.

Since you change the argument to base it on the least significant cost a
business faces, and pretend that there are no other costs to the business, you
present an utterly fallacious case.

Straw man. I did not pretend that there were no other costs. I glossed
over the fixed costs because they'll be paid off in finite time and
after that the graph climbs into the black and just keeps climbing. As
long as their price exceeds their net ongoing costs per unit. (This
includes amortized support costs if they don't separately charge money
for support, and otherwise it's just the incremental cost of stamping
out one more plastic disc or pushing a few more electrons down a
wire.)

You would excoriate the farmer for charging fair prices for produce because
their farmstand costs so little to set up. What about the whole growing
season of labor, when they're spending and spending and spending hoping their
crops will recoup the costs? Then you come along and tell them they're
monopolists because the "marginal cost" of a farmstand doesn't justify their
prices.

That doesn't make any sense. They have a whole lot of labor per ear of
corn or whatever, which they charge for. This is very different from
doing a whole lot of labor once in the farm's entire lifetime and then
sitting back and watching it grow all by itself unattended and still
charging just as high a price for it. Of course farmers aren't that
lucky, but even so, the one-time cost of buying the land is probably
not figured into the price in any way; only the per-ear cost of
raising crops of corn after the land is theirs (and any ongoing land
taxes they might have to pay, divided by the number of ears in each
harvest).

You'd have all software vendors out of business in no time.

By your theory, Red Hat should be wallowing in debt. Explain the
observation that it is not.

Remember for future reference, "marginal cost of distribution" is not the
only, nor even the most significant cost of software for the manufacturer.

It is in the long term, aside from support, which they can always
charge separately for (as Red Hat does).

 

[nebulous99]

The thing you're forgetting it the cost of creating the software,
database, collated information or whatever. Unless you're a hobbyist
that cost needs to be recouped.

It's a one-time cost though.

There are a few ways:

4) Spread cost+profit over a projected sales target, 'n' and charge
each customer (cost+profit)/n - the usual sales model BUT it
depends on preventing free copies being made and distributed.

Not really. It means you'll hit the target later, but you can still
hit it. You can compete with free. Also, once the target is hit
there's no justification for not dropping the price to just above
marginal anymore OR for disallowing copying.

A business model that depends on regulating heavily the downstream use
of the thing after you sell it is a poor choice for a variety of
reasons anyway.

6) Err, I don't think there is a 6, at least not if you want to
earn a living from your work.

Non sequitur. An individual "earns a living from their work"; a
corporation "makes a profit". Which are you now talking about? An
individual with proven programming talent should be able to find work
coding, no matter how profits are earned by their employers. Which
might be selling hardware widgets, and the employee works on widget on-
board software, or on the office's accounting systems administration,
or who knows what.

As for a proper number 6, how about

6) The development costs are shared with a wider community, which is
made possible by not being proprietary, and afterward you find a way
to compete with free, and turn a profit and even become a stock market
darling in the process. AKA the Red Hat business model.

Given that 6 obviously can work, why should society do anything to
help the brutish 4-using vendors enforce their wishes arbitrarily on
third parties long after they've concluded their transactions with
said vendors? There's obviously no need, and kicking the crutches out
from under those nasty vendors would force them to switch to something
nicer that still works, like 6, or go under. (If there's need for
whatever their product was, someone else will find a way to market it
successfully by being smarter and more agile since there's demand to
be met. And the dead vendor's employees with coding talent should be
able to find work, perhaps at such a replacement, but certainly
somewhere in the programming field anyway.)

 

[Lew]

No. It is you who are wrong. You conveniently snipped the rest: "They
could sell copies for a
dime apiece and turn a profit in the long run."

Then let me address the point. That is a false assertion. They likely would
show a loss in the short and in the long run.

The "other costs" you speak of are one-time costs. They spend these
once, then the marginal cost for each copy sold.

They still have to recoup the costs.

So they're initially in the hole $D, the development one-time costs. Subsequently, they get
$P-$M, price minus marginal cost, on each copy. Assuming $P-$M is
positive, the amount they're in the hole by shrinks linearly over time
and eventually goes negative. At that point they're showing a net
profit and from then on it's gravy train time.

"Eventually" being a very long time. Anyway, what is your beef with profit?
Profit is a good thing. Providers should make the maximum profit the market
will allow.

It's elementary arithmetic. That you cannot grasp this shows that you
are not a fit opponent for this debate. Stop posting over and over
again arguing with me and forcing me to waste my time refuting your
drivel and unwanted, mathematically-bankrupt opinions.

I force you to do nothing, except perhaps to resort to insults when logic is
unable to serve your argument.

Straw man. I did not pretend that there were no other costs. I glossed
over the fixed costs because they'll be paid off in finite time and

Same difference. Calling it "straw man" doesn't make it so.

after that the graph climbs into the black and just keeps climbing.

Your scheme would ensure that it never does that.

As long as their price exceeds their net ongoing costs per unit. (This
includes amortized support costs if they don't separately charge money
for support, and otherwise it's just the incremental cost of stamping
out one more plastic disc or pushing a few more electrons down a
wire.)


That doesn't make any sense. They have a whole lot of labor per ear of
corn or whatever, which they charge for. This is very different from

Actually, it's exactly the same. The developer also has "a whole lot of labor
per [functional unit of software] or whatever", which they have to invest
first before they can make a sale.

doing a whole lot of labor once in the farm's entire lifetime and then
sitting back and watching it grow all by itself unattended and still
charging just as high a price for it. Of course farmers aren't that

You have a complete misconception of how it works.

lucky, but even so, the one-time cost of buying the land is probably
not figured into the price in any way; only the per-ear cost of
raising crops of corn after the land is theirs (and any ongoing land
taxes they might have to pay, divided by the number of ears in each
harvest).

Calling development costs "one-time" costs doesn't make them so. You
conveniently glossed over all the other costs, again, such as overhead, labor,
marketing, ... You can't just ignore reality. Of course, you obviously don't
run a business yourself or you'd understand such elementary facts.

Or maybe you do, and you're just being contrary so you can impress yourself
with how clever you imagine your arguments to be.

By your theory, Red Hat should be wallowing in debt. Explain the
observation that it is not.
Rrr?


It is in the long term, aside from support, which they can always
charge separately for (as Red Hat does).

Bullshit. You don't have a clue.

 

[Twisted]

Then let me address the point. That is a false assertion. They likely would
show a loss in the short and in the long run.

If you are this bad at math, then you should just shut up now and stop
constantly posting attack responses to everything I write.

I will spell it out for you now, in extremely simple language that
even a complete and utter moron can understand. If you still fail to
understand, then you would be going into my killfile if fucking Google
Groups had such a feature, and will definitely no longer get detailed
responses from me, since any such response will obviously just sail
right over your head.

Q: Little Johnny spends $10 to develop a piece of software and then it
costs $0.05 for him to make each copy. He sells copies at $0.10 each.
How many copies must he sell to break even? To have $10 more than he
began with?

A: Each time he makes and sells one copy, it costs him a nickel to
make and he receives a dime, so he gains $0.05. Two copies sold nets
him $0.10 in revenue. Two hundred nets him $10 and he breaks even when
he sells the 200th copy. Every copy he sells after that is a nickel of
profit. After the 400th copy he has $10 more than he began with. After
the 1000th copy he has forty whole dollars to show for his enterprise.

They still have to recoup the costs.

See above. Little Johnny recoups the costs after selling 200 copies.

"Eventually" being a very long time. Anyway, what is your beef with profit?
Profit is a good thing. Providers should make the maximum profit the market
will allow.

Yes, but a properly functioning market should also hold them down to
just above marginal cost. Thin margins are a sign of healthy
competition. Fat margins (especially as obese as each sale being 99.9%
net revenue!) indicate a serious problem has developed in that
department. Healthy competition is more "just", because things are as
affordable as they can be given the actual costs of manufacture,
making it possible for the poor to get something that costs $10 to
make if they have around $11, instead of it being priced out of their
reach at $300 or something. Healthy competition is also more
efficient, and the economy grows much faster than if monopolists can
just stagnate and rake in money without doing very much except the odd
lobbying or suing here and there.

I force you to do nothing, except perhaps to resort to insults when logic is
unable to serve your argument.

You are posting total BS, which needs to be publicly debunked in case
someone otherwise actually reads and believes the nonsense you wrote.
See the above math problem with Little Johnny, which proves that you
are wrong. Now will you please give up trying to "prove" your
cockamamie theory right when it's been roundly mathematically
disproven? It can't get any more disproven than that!

Same difference. Calling it "straw man" doesn't make it so.

It is a straw man argument. One-time costs do not affect long-term
profitability. Recurring costs are the ones to watch out for.

Your scheme would ensure that it never does that.

No, it wouldn't. Little Johnny is in the hole $10 for a while but
eventually is showing a profit, in case you already forgot. All
businesses have to invest some money in R&D before they can make it
back and then keep making additional money with a product. This is
normal. It is in no way unique to the software industry. GM may spend
millions designing a new car model, which is based on existing
technology so no patent protection for that investment. If they make
hundreds on each car sold, as seems likely, then after the first ten
thousand units or so shipped they've broken even and every subsequent
car of that model sold adds to their coffers. Where do they get the
millions? From the profits from earlier operations. A startup would
get it from venture capital or investors of some other sort.

Movie studios spend enormous amounts to make a movie. Sometimes they
make back their production costs in the theatre and sometimes they
don't. Enough do that they turn a profit on average, even if sometimes
they're temporarily in the hole by tens of millions due to an
expensive flop like Waterworld. Note that this is before the heavily
DRM-encumbered DVDs launch; usually a box-office failure doesn't make
much more in video sales anyway. I think the studio that made
Waterworld still hasn't turned a profit on it after a couple decades
of VHS and DVD sales, and all the DRM and copyright in the world
didn't do any good there. It simply wasn't popular. If it had been,
all the free copying in the world wouldn't have STOPPED them turning a
profit. It might have slowed it down. It might actually have
accelerated it by free copies generating additional buzz and through
sales of merchandise of various kinds -- popularity would have meant
being able to sell mugs, T-shirts, and action figures by the ton-lot.

Flops will still be flops. Hits will still be hits. Only the way the
money is made off the hits might change. There isn't much to be made
off the flops, and there never will be, unless you outlawed freedom of
speech and outlawed bad reviews and imposed amazing levels of police-
state nastiness to quell word-of-mouth "reviews" from spreading. We
certainly don't want to do THAT just to protect a few special
industries' profits, ones that are profitable anyway. Why in fact do
any curtailing of freedoms for such a purpose then?

Actually, it's exactly the same. The developer also has "a whole lot of labor
per [functional unit of software] or whatever", which they have to invest
first before they can make a sale.

No. The development costs are like the farmer's land. They are paid
once and then copy after copy can be made and sold for cheap. The
farmer isn't so lucky. After buying the land and paying that cost
once, ear after ear of corn can be grown but it's rather more labor
intensive per ear of corn, and there's probably land taxes too to
amortize over each year's crop.

You seem to be laboring under the delusion that the development cost
for a piece of software has to be paid again for every single copy,
which is utter hogwash. Once a single copy has been made at whatever
cost, further copies are pennies each -- the cost of stamping one more
CD at a factory somewhere. It goes up to a buck or so if the volume is
low enough, but even then no higher -- I can burn CDs one at a time
for about a buck each myself, so the "small press" limit of the
marginal cost of reproduction of software is around $1 per 650MB. For
very big chunks of software that's potentially as much as $5. With
high volume and a disc-pressing factory it's still peanuts -- nickle-
and-dime territory. Adding shipping and handling costs can raise it to
a buck a disc or so in volume, and maybe $20 a disc "small press".
Downloading instead of making physical discs brings it back down to
pennies. It's certainly never anywhere near $300.

Remember that, assuming they don't provide free support, these are the
ONLY costs that grow proportionally to the number of copies they make
and sell. And those costs are not raised at all by any third-party
copying and distribution.

If the purpose of selling copies is to pay down the larger cost of
making that very first copy out of whole cloth, selling them for more
than around $40 (discs in small lots) to $0.25 or so (downloads)
suffices to ensure your initial costs get paid down.

The only real complication is that if the initial costs were paid out
of borrowed money there's interest, and prices need to increase to pay
down the debt faster than interest accrues so that it eventually
disappears instead of growing. In this case, the added amount needs to
be higher for lower sales volume, and lower for higher sales volume.
For very high sales volumes (e.g. Windows XP when it launched) you can
still charge say a buck a download and pay off any reasonable
combination of debt and interest-rate in a few years -- especially
since lowering the price further increases sales volume!

And there's also the option of saving up money to make the initial
investment instead of using a loan to finance initial development.
Then there's no interest to worry about when paying off the initial
costs. Although you probably still want that "principal" to be paid
off at a rate exceeding inflation of course.

Calling development costs "one-time" costs doesn't make them so. You
conveniently glossed over all the other costs, again, such as overhead, labor,
marketing

You're inventing this claim out of whole cloth.

What goes into making a piece of software like say Windows XP?
You make a first copy. This costs lots.
You make more copies, which each cost very little.

The one large cost happens once, up-front. It does not recur, unless
you decide to make something like Windows Vista, which is obviously a
gamble (and MS looks like it might lose this one).

Overhead, labor, and the like associated with the development effort
is part of that cost-of-the-first-copy. Overhead, labor, and the like
associated with each disc pressed subsequently is part of the marginal
cost of reproduction, and we are assuming the price per copy is a bit
higher than that cost.

Marketing is optional. Spending a bunch of money to spam everybody in
sight is a waste of money these days, when there's Internet word of
mouth. Make something good and people will find out about it and beat
a path to your door. With something like a new version of Windows, a
single bit of public fanfare should start it selling at a decent rate,
and people won't need reminding, because they'll be saturated in
plenty of mentions of the product. I avoid and block Internet ads and
aren't watching much TV during summer rerun series, but I run into
mentions of Vista constantly. No, I don't need reminding that it's out
there and if I didn't think it was at best a waste of time and money
I'd be getting a copy or preparing to without any extra prodding from
Microsoft.

Marketing is only really needed to generate sales of a product that's
shoddy and nobody much wants, or something that's really obscure, and
in the former case I say tough shit, and in the latter case the
marketing should be more narrowly targeted.

And marketing is finely tuned. Any sensible company will figure out
that if they spend $X a year on marketing and use it in various ways,
they'll get $Y a year worth of extra sales versus if they spent
nothing, and look for where $Y is as much bigger than $X as possible.
In that case, every dollar spent on marketing gives rise to an extra
$1.10 or whatever in revenue, so marketing is actually a profit center
when you think about it, and is paying for its own costs.

Stupid companies of course spend $7 zillion on saturation-bombing the
planet with their spam and end up making $40,000 worth of extra sales
and then wonder why they are not doing well on the stock market. These
ones can go straight to hell for all I care; it's their own damn fault
for not doing the math.

Rrr?

They let anyone make and give away or even sell copies of their
flagship software product. They sell copies themselves at around $60 a
pop. They make money on that because the copies are so cheap. They
also sell paid support services, and finance further development out
of the profits from the support services (and also outsource it,
partly to volunteers at no cost). It works -- everyone seems to come
out a winner. Red Hat, its competitors, its customers, and its
freeloading non-customers. You cannot possibly contemplate arguing
against that kind of success!

(Also notable is that the support service presumably generates
information guiding further development, as well as the money to
finance the costs of developing new versions.)

[Remaining rudeness deleted]

Lew, feel free to shut up now. You clearly don't know what you're
talking about, and you clearly have all the mathematical intuition of
a potted petunia. Figuring out if the revenues exceed the costs long-
term and doing this accurately is clearly not something you're capable
of achieving, and you can't possibly make a cogent and correct
argument on this topic if that is the case. Certainly your attempts so
far have been dismal failures. So I kindly suggest you cut your losses
and stick to posting about Java itself.