Learning C++

[Balog Pal]

Either that, or the ten thousand features vi has offered for 30 years are
actually useful.

It is similar to std::string. Indeed vi has a feature other stuff lacks: it
is *there*. Out of the box. Back in time some of my friends learned to use
edlin of DOS. For that same purpose. Because going around they could edit a
file on any random computer.

Btw it is just wild luck vi got in -- we could have the same conversation
you protecting ed. And there probably were similar debates at the proper
time point why vi is needless.

>There's a reason people still use vi and emacs, why mail
servers still use sendmail, and why people still use RSA and ethernet and
binary computers.

The usual shifting of context. "Use" is fundamentally different to "Use for
XXXX purpose". vi is certainly useful for the maitainance guy to fix the
..profile on a random computer. It in no way imply that it is also fit
for professional software development where being effective and accurate is
on the target list.

Suggesting any of those to be "youthful rebellion" is purely nonsense.

And I did not claim sensible usage of "tool for the purpose" like that.

However claiming tha raw general tool being superior to the one developed
with the clear purpose -- and btw used in greater numbers around the globe
too, just by people who are not loud in forums -- sounds like that.

With all the "supporting" other statements

 

[Rui Maciel]

You're right - it doesn't. Fortunately for me, that's not what I wrote.

I have to say you are right, I've completely misread what you wrote. Sorry
about that.

IDEs are "incredibly complex"?!?!? Wow, speaking of "astonishingly
silly" notions. Have you considered that, to someone who's 20+ years
younger than you or I, GUIs are the *normal* way they've been doing
things for as long as they've been using a computer?

Just because you slap a GUI onto a tool it doesn't mean it becomes simple (or
even simpler) to master. IDEs are packed with lots of features and, in order to
make them accessible to a user, they need to include ungodly amounts of menus,
buttons and dialogs. If a user is placed in front of an IDE he isn't familiar
with then he is faced with a sea of options that, at least at first, he won't be
able to navigate through without taking his fair share of time to think things
through, even being forced to rely on trial and error to find what he is
searching for.

Now, you place a newbie in front of that same IDE. The newbie's difficulties
are even greater because not only he isn't familiar with that IDE but he also
doesn't have a clue about what he is supposed to do to begin with.

On the other hand, that becomes a non-issue if the newbie doesn't make the
mistake of jumping on the IDE bandwagon and simply relies on a text editor to
write his code. After all, there's a reason people pay good money for IDE
courses and you don't see any of those scams being done with text editors.

Please note - I'm not saying that no one should ever learn command-
line compilers and make files. I'm just saying that, on the first week
of class, learning to program is difficult enough; allowing students to
use a familiar graphical environment that resembles the applications
they've grown up with will be less jarring for them.

....which, as you will be hard-pressed to find a single text editor which is not
a GUI application (and let alone force anyone to use it), may be applied to any
case.

Yet, it is far better to let the newbie compile things manually. If he falls
under the impression that "compiling" means pressing that funny looking button
then, without even mentioning the rest of the abstractions IDEs impose on their
users, you will end up with a "programmer" who doesn't even have the faintest
clue about how his code manages to be turned into a binary. It will all be
magic to them. And that isn't in the newbie's best interests.


Rui Maciel

 

[Andrew Poelstra]

It is similar to std::string. Indeed vi has a feature other stuff lacks: it
is *there*. Out of the box. Back in time some of my friends learned to use
edlin of DOS. For that same purpose. Because going around they could edit a
file on any random computer.

This is true - and it's actually the biggest reason that I do use vi.
Because I can ssh home from any system, anywhere, or onto any system
for that mattter, and it's there and it works. I'm using it right now
to compose this message.

Btw it is just wild luck vi got in -- we could have the same conversation
you protecting ed. And there probably were similar debates at the proper
time point why vi is needless.

Well, I would submit that not only was vi designed for professional
development, it does a mighty fine job. But that's a debate I don't
want to get into.

 

[Brian]

A lot of us don't use windows.  While there are good cross platform
IDEs, they are fine for green field hosted development but they don't
support cross-compiling for non-native targets.

When a project mixes code from different sources (open source libraries
for example), makefiles are usually the best, if not only, option.
Every project I work on has a mix of local and open source code, so the
build system has to cope with both.  There's so much open source code
out there, the sooner a developer gets to know how to build form the
command line, the better.

The following comments are from the review of the new
book "You Are Not A Gadget" in the Wall Street Journal --

http://online.wsj.com/article/SB100...74646402192953052.html?mod=article-outset-box

"There's a dominant dogma in the online culture of the moment
that collectives make the best stuff, but it hasn't proven to
be true. The most sophisticated, influential and lucrative
examples of computer code—like the page-rank algorithms in the
top search engines or Adobe's Flash— always turn out to be the
results of proprietary development. Indeed, the adored iPhone
came out of what many regard as the most closed, tyrannically
managed software-development shop on Earth."


(I suspect there's some truth to that tyrannical comment
and thank G-d I don't need to seek employment with Apple.)

There's no denying there's tons of open source out there,
but so much of it is of low quality. In case anyone needs
another example of that --
http://webEbenezer.net/comparison.html.


Brian Wood
http://webEbenezer.net
(651) 251-9384

 

[Ian Collins]

The following comments are from the review of the new
book "You Are Not A Gadget" in the Wall Street Journal --

http://online.wsj.com/article/SB100...74646402192953052.html?mod=article-outset-box

"There's a dominant dogma in the online culture of the moment
that collectives make the best stuff, but it hasn't proven to
be true. The most sophisticated, influential and lucrative
examples of computer code—like the page-rank algorithms in the
top search engines or Adobe's Flash— always turn out to be the
results of proprietary development. Indeed, the adored iPhone
came out of what many regard as the most closed, tyrannically
managed software-development shop on Earth."

There's no denying there's tons of open source out there,
but so much of it is of low quality.

There are also a lot of high quality, long standing opens source
projects as well. The products my team develops use several and they
are a lot better quality than a large piece of smelly proprietary
software we have to use.

 

[Jerry Coffin]

(e-mail address removed)>, (e-mail address removed)
says...

[ ... ]

Search and replace, for starters. You still have to enter the
regular expression and the replacement text. By the time you've
gotten the dialog box open and your hands back to the keyboard,
I've finished the command.

With VS, opening the dialog only takes a ^h -- no need to remove
hands from the keyboard at all. If you prefer a different key
combination, you can reassign it (actually, I should add that I'm not
sure control-h is really the default, but that's what I have it set
to, and at least offhand I don't recall having changed it).

The one thing about search and replace that IS really annoying is
that they require you to hit the tab key _twice_ to move from the
"search for" to the "replace with" fields in the dialog. You get used
to that fairly quickly, but it still strikes me as a poor design.

When editing (entering text), anything which requires using the
mouse is a minus, since it requires taking your hands away from
the base position on the keyboard. (When just browsing, of
course, other rules hold, and I do tend to use the mouse for
most browsing.)

At least with the VS IDE, you can assign virtually any command on any
menu to a key-combination of your choice. Quite a few of them already
have keys assigned, of course, but you can change the assignments at
will -- though I'd admit that it borders on ironic that (at least by
default) no key combination is assigned to open the dialog to do key
assignments, so you need to select Tools | Customize, then click the
"keyboard..." button at the bottom left of the dialog that shows up
(at least the first time -- though this is something you do rarely
enough that I've never seen a reason to assign a key-combo to do it).

Here again we find a bit of poor design: I've never tried to count
them, but at a guess, there are at least a few hundred commands you
can assign to keys of your choice -- but on the dialog they're
displayed in a tiny list box that will only show about four at a
time. Just finding a particular command can be pretty painful.

One frequent operation for me is grepping for every place where
some function is used. Then invoking the editor on just those
files, running through the first modification, memorizing the
keystrokes in a macro as I do, then undoing it, and invoking it
on all of the files I'm editing (and only on those files---with
a bit of care in my original search string, I can only modify
specific uses of my function).

Control-shift-f opens a "find in files" that supports regular
expressions. That puts the names of the files in a "find results"
window. Control-tab to select that as the current window. In that
window, control-a to select all, then Return to open all the files.
Control-shift-r to record your macro. Do your editing, then control-
shift-r again to stop recording. Control-z to undo the macro run on
that file, then Control-shift-p to run it. Control-s to save that
file, and control-F4 to close it.

If you really do that a lot, I'm pretty sure you can create a macro
to automate it so about all you'd do is one key combo to run the
macro, type in your RE, and do the editing you're going to do in your
macro. It can handle everything from there... The only part I'm not
sure about is having one macro record another macro -- I'm pretty
sure you can probably do that, but I've never tried, so I can't say
with real certainty.

 

[dragan]

A friend of mine is moving to California and did not want to move all
his library.

"Brokeback Mountain"?

 

[gwowen]

There's no denying there's tons of open source out there,
but so much of it is of low quality.  

In what regard is that different from closed source software?
Or music, film, books or food?
Or, for that matter, politicians, athletes, priests, computer
programmers or traffic wardens?

 

[James Kanze]

There are also a lot of high quality, long standing opens
source projects as well. The products my team develops use
several and they are a lot better quality than a large piece
of smelly proprietary software we have to use.

I don't know about lots; there are some reasonably good open
source projects, even if they're far from the majority. And a
lot of closed source is pretty poor as well. (Of course, when
the projects aren't open source, it's hard to know whether
they're good, or just lucky.) The very best software (in terms
of reliability) generally comes from more or less rigorously
organized environments, which puts most freeware at a
disadvantage, but a lot of commercial firms aren't that well
organized either. In the end, if you are purely interested in
statistics, all it takes for an open source to survive is
stubbornness on the part of the author---a commericial project
which is really, really bad finishes by going out of business.
But does it make sense to include all of the open software that
no one uses in the statistics?

In the end, if you need a program to do something, evaluate it
on its merits. Without letting open source or not influence
your judgement.

 

[Ian Collins]

I don't know about lots; there are some reasonably good open
source projects, even if they're far from the majority.

I never said they were the majority. But everything on my home and work
desktops and servers is open source (OpenSolaris and Linux) and quality
wise it's up with any closed source alternative. The products my
company ships are based on a plethora of opensource packages and build
with gcc plus the one large piece of smelly proprietary software. We
have the source for the latter, so we know how bad it can be!

The very best software (in terms
of reliability) generally comes from more or less rigorously
organized environments, which puts most freeware at a
disadvantage, but a lot of commercial firms aren't that well
organized either.

Based on my fairly extensive experience, the organisation of open source
projects and commercial firms is pretty similar. There is a similar mix
of chaos and order. It comes down the the individual(s) who drive the
projects.

In the end, if you need a program to do something, evaluate it
on its merits. Without letting open source or not influence
your judgement.

I basically agree with that, although the issue is more complex (did
anyone mention licensing?) when one is evaluation a piece of software to
be included within a bigger product.

 

[Brian]

I don't know about lots; there are some reasonably good open
source projects, even if they're far from the majority.  

I thought about challenging the "a lot" also. The only
open source project that uses C that I like is bzip2.
I use linux and gcc, but they both have caused me some
grief. Boost and Loki I think are two C++ based, open
source projects that I think sometimes do pretty well.
There are so many others that don't do as well. Recently
I've been looking at openssl. That is really disappointing.
I tried searching for a C++ alternative but haven't found
anything. There's something called crypto++ but it doesn't
seem to have an ssl implementation. I sent an email to the
author and he said he was too busy and referred me to the
mailing list. The answer on the mailing list was to
consider using openssl. I am reconsidering openssl, but
only because beggars can't be choosers or something like
that.

And a
lot of closed source is pretty poor as well.  (Of course, when
the projects aren't open source, it's hard to know whether
they're good, or just lucky.)  

Well, some companies allow you to download their software
for a free trial. Others have free on line services that
you can evaluate at your leisure. So the "hard to know"
is maybe more like it takes a little effort to figure out
if they're good or not.


Brian Wood
http://webEbenezer.net
(651) 251-9384

 

[Rui Maciel]

James Kanze wrote:

In the end, if you are purely interested in
statistics, all it takes for an open source to survive is
stubbornness on the part of the author---a commericial project
which is really, really bad finishes by going out of business.

Not necessarily. There are commercial software packages exist for quite a while
and although they have their fair share of weaknesses they are still here and don't
appear to go away. Possibly it's just due to lack of competition, serving niche
markets and requiring quite a lot of know-how but they do exist.


Rui Maciel

 

[tanix]

There is no rule one way or the other.
It all depends on awareness of individuals.

I doubt it. Haven't seen a single one to date.

The only advantage I see is the abscence of commercial pressure.
So, it is more relaxed environment. But that often means
sloppiness.

Well, the proprietary code IS smelly in that in vast majority
of cases it is non documented and written in such a way as to
make it virtually unreadable by someone else. Job security issue.

I don't know about lots; there are some reasonably good open
source projects,

I wish I saw one.

even if they're far from the majority. And a
lot of closed source is pretty poor as well.

Indeed. Except because of typical viciousness in the sw
business, it tends to be much less documented and much less
readable.

(Of course, when
the projects aren't open source, it's hard to know whether
they're good, or just lucky.) The very best software (in terms
of reliability) generally comes from more or less rigorously
organized environments,

Well, it immediately associates with viciousness in my mind.
The most disgusting environments I had to deal with use this
"rigorous" approach, which turns out to be a living hell.

There is a subtle balance between precision and humaneness.
Long subject.

which puts most freeware at a disadvantage,

With freeware people somehow think that the only think they have
to do is to write there "latest and greatest", "revolutionary"
piece of code and documentation and clarify part of it is not
their concern. Most of that code ends up being unreadable
without wasting days on trying to understand their densely
packed piles of totally undocumented crap.

but a lot of commercial firms aren't that well organized either.

Well, when you are too much concerned with the "bottom line",
you have not much time left and no incentive for people to
take it to the level of elegance and document it so it would be
a breeze to read. Just the other way around, the more unreadable,
the more undocumented your code, the better is it for your
"job security". Sick stuff basically.

In the end, if you are purely interested in statistics,

Which is meaningless.

all it takes for an open source to survive is
stubbornness on the part of the author
True.

---a commericial project
which is really, really bad finishes by going out of business.
But does it make sense to include all of the open software that
no one uses in the statistics?

Well, the open source gives you at least one consolations:
if something does not work as you expect it, you can change it
and make your own verions. Except you never seem to have enough
time to even BEGIN to read all those piles of totally undocumented
spaghetti code. So it does not work out at the end.
But at least you do have the option.

In the end, if you need a program to do something, evaluate it
on its merits. Without letting open source or not influence
your judgement.

Well, all equal, i do prefer the open source version.
At least you are not locked in into some dead end situation.

I think the major problem with sw business is the pricing issue.
If companies like microsoft would not set a precedence of charging
an arm and a leg for their stuff, it would be much more beneficial
to the whole industry.

Sure, everybody wants a "free ride". But most people can afford
to pay some small amount of money to be able to play with some
fancy sw toy. Say like $50 for someting that really gives you some
joy for a least a couple of months.

The idea of "open source" has its limitations in that those,
who spend a considerable amount of time to write some piece of
code, are forced to simply give it away for free. And to whom?
To some parasites that are forever looking for a "free lunch"?

If you spend years of your life writing some piece of code,
why aren't you compensated for your effort? Are any of those
parasites, using your code, are willing to dedicate years of
THEIR lives to do the same? Or are they simply interested in
sucking?

Basically, people need to pay more attention to the fact that
you have to pay something to get something of value. You need to
be greatful to those, that gave you some joy by writing such a
beautiful piece of code.

And if you are NOT greatful, you are ripping YOURSELF,
and not the other as you might think.

That's the verdict.

--
Programmer's Goldmine collections:

http://preciseinfo.org

Tens of thousands of code examples and expert discussions on
C++, MFC, VC, ATL, STL, templates, Java, Python, Javascript, PHP,
organized by major topics of language, tools, methods, techniques.

 

[tanix]

I never said they were the majority. But everything on my home and work
desktops and servers is open source (OpenSolaris and Linux) and quality
wise it's up with any closed source alternative. The products my
company ships are based on a plethora of opensource packages and build
with gcc plus the one large piece of smelly proprietary software. We
have the source for the latter, so we know how bad it can be!


Based on my fairly extensive experience, the organisation of open source
projects and commercial firms is pretty similar. There is a similar mix
of chaos and order. It comes down the the individual(s) who drive the
projects.

Agreed. The "environment" factor.
But it is not easy to maintain the healthy environment in the
commercial situation, where you are forever stressed out to the hilt.

I basically agree with that, although the issue is more complex (did
anyone mention licensing?) when one is evaluation a piece of software to
be included within a bigger product.

To my opinion, it is time to reconsider all the major principles
of "doing business" and "licensing" is one of them.
The models that exist right now are not based on principles of life.
They are based on principles of sucking. As much as you can manage.
As a result, we have the whole planet in the miserable state it is in.

--
Programmer's Goldmine collections:

http://preciseinfo.org

Tens of thousands of code examples and expert discussions on
C++, MFC, VC, ATL, STL, templates, Java, Python, Javascript, PHP,
organized by major topics of language, tools, methods, techniques.

 

[tanix]

I thought about challenging the "a lot" also. The only
open source project that uses C that I like is bzip2.
I use linux and gcc, but they both have caused me some
grief. Boost and Loki I think are two C++ based, open
source projects that I think sometimes do pretty well.
There are so many others that don't do as well. Recently
I've been looking at openssl. That is really disappointing.
I tried searching for a C++ alternative but haven't found
anything. There's something called crypto++ but it doesn't
seem to have an ssl implementation. I sent an email to the
author and he said he was too busy and referred me to the
mailing list. The answer on the mailing list was to
consider using openssl. I am reconsidering openssl, but
only because beggars can't be choosers or something like
that.

Well, what I see is that people somehow underestimate the
value of something added to something.

Look at it this way. The guy(s) wrote some piece of code.
It works. It does something of value to you.
But...
As it oftent happens, it is poorly documented and quite often
poorly structured. Not necessarily a good architecture.

But...

That is ALL that is avaiable to you.
Something HAS been done.
So, what are your options?

Well, either you live with it as it is, or...

Tadaaam!

Add something of value to it and give it another spin.

If the value you have added is available to others, then
they can do the same thing.

The basic principle is care-ness. You need to CARE.
If you care, it is life affirmating.
It ADDS to the equasion.

The very radical idea of "free software" or "open source"
has basically to do with this.

Yes, it takes some inner strength to be able to afford to give
something to others without getting a penny in return.
It kinda contradicts the whole "modern world" ideology.

But...

It is not such a "stoopid" idea, just as open source approach
shows. It is alive. It does produce value. And millions of people
enjoy that value and enjoy that very feeling of not being tied
up with their hands to some mega-sucking enterprise, such as
microsoft, google or poogle.

It IS a life affirmative approach.

Well, some companies allow you to download their software
for a free trial. Others have free on line services that
you can evaluate at your leisure. So the "hard to know"
is maybe more like it takes a little effort to figure out
if they're good or not.

I bet it would be more beneficial to reconsider the pricing issue.

Sure, free trial gives you at least something.
You don't have to shell out an arm and a leg just to see if
you like something enough to get married to it for a while.

But at the same time, one way or the other, you need to pay for
things. You need to pay for water. You need to pay for electricity.
You need to pay for food.

In today's world, software is like water or food.
And you can not expect to get something for nothing.
I would rather see Sun charging a small fee for Java
instead of giving it for free and eventually going out of business,
being bought by some sharks of Oracle, Microsoft of Google type.

You don't have to go to extremes.
Either you charge an arm and a leg or give it for free.

Just look at how they managed the issue with television in UK.
Everybody simply pays some annual fee and is guaranteed to have
a more or less unbiased television. What is wrong with THAT kind
of approach? After all, there seems to be no rebellion of people.

The same kind of thing could be done with software,
and it would be MORE beneficial to whole life at the end.
And I bet there are quite a few ways to solve this situation
so everyone gets a benefit and at the same time realizes there
is price for everything. But, as long as it is reasonable and
afofdable to ALL, regarless of how big is their wallet, it should
be no problem to even mention.

If you are interested in playing with ideas of writing some code
and that code turns out to be interesting enough to others,
why can't you get paid for it enough for you to get some bread
on your table? Sure, if you have great ambitions, you are welcome
to enter the land of sharks of Microsoft caliber and pay an arm
and a leg for using their "revolutionary technology" of suckology.
Because you are going to get some tangible financial benefit out
of it. But that is a different story.

Interesting thing about sofware is that it is essentially a toy
in most cases.

There are very few pieces of software you use every single day
as a "professional occupation". If you do, that is another matter.
You should pay a fair share of benefit you get from using it.

But MOST of what we use are toys. Those things that give you fun
and provide challenge to your mind to move into new dimensions.

I would not mind to pay some reasonable fee for using Linux.
I don't have to get it for "free". That is just illusion.
PLENTY of people contibuted to it. Why can't we find a way
for them to get paid?

Basically, what I see is two different categories:
1) Commerical use, when you expect to get a tangible financial
benefit out of it.

2) Non commercial use. Such as self-education and fun.

Those are totally different issues and different principles
ought to apply.

To have a reasonably priced software of ANY kind you'd like to
play with will benefit the mankind as a whole and will help to
solve the most urgent issues we have on our hands.

If I can have ANY kind of software I can imagine, that would
stimulate me to play with new and different things. Who knows,
may be one of these days I will find something I am willing
to spend a significant enough time on, which will inevitably
translate into some value.

Brian Wood
http://webEbenezer.net
(651) 251-9384

--
Programmer's Goldmine collections:

http://preciseinfo.org

Tens of thousands of code examples and expert discussions on
C++, MFC, VC, ATL, STL, templates, Java, Python, Javascript, PHP,
organized by major topics of language, tools, methods, techniques.

 

[Ian Collins]

I wish I saw one.

Pretty much everything in the web stack: Apache, MySQL/PostgreSQL, PHP,
your open source OS of choice and MediaWiki to document it.

 

[Brian]

In what regard is that different from closed source software?

I think closed source software is generally with some sort
of company that is seeking to make a profit from their
investment in creating the software. Open source projecta
are sometimes nothing more than a negative reaction companies
making money from their products. If that is the primary
basis of an open source project, I think they tend to not
be very good. If generosity is the basis of an open source
project, I think it tends to be better. Some open source
projects are probably a mixture of motivations. One thing's
for sure, the difficult nature of the times is testing the
foundations of a lot of companies and projects. We are,
more than ever, in a survival of the fittest time. I'm
no Darwinian though.

Or, for that matter, politicians, athletes, priests, computer
programmers or traffic wardens?

What's a traffic warden? Speaking of politicians, I liked
an article by Frank Rosenbloom about Nancy Pelosi titled,
The Bad Lady Flies a Boeing 757. http://frankrosenbloom.com


Brian Wood
http://webEbenezer.net
(651) 251-9384

 

[nick]

Pretty much everything in the web stack: Apache, MySQL/PostgreSQL, PHP,
your open source OS of choice and MediaWiki to document it.

Not to mention gecko and webkit... at least second best to the closed-
source alternatives, if not better. Drupal is good enough for
whitehouse.gov.

And then there's cmake, and so many great code libraries are published
under non-restrictive (LGPL-like) open-source licences (qt, boost,
zlib, png, the list goes on and on).

Plus, many of the leading software architecture tools offer both
commercial and open-source licenses, and some of the commercial tools
are built on open-source tools -- for example ArgoUML / Poseidon.

In other words, I don't see how there's any way anyone _hasn't_ seen,
used, or at least benefited unknowingly from some "reasonably good
open source projects" unless they've been living under a rock for the
last 20 years.

-- Nick

 

[Brian Wood]

Well, what I see is that people somehow underestimate the
value of something added to something.

Look at it this way. The guy(s) wrote some piece of code.
It works. It does something of value to you.
But...
As it oftent happens, it is poorly documented and quite often
poorly structured. Not necessarily a good architecture.

But...

That is ALL that is avaiable to you.
Something HAS been done.
So, what are your options?

Well, either you live with it as it is, or...

Tadaaam!

Add something of value to it and give it another spin.

If the value you have added is available to others, then
they can do the same thing.

The basic principle is care-ness. You need to CARE.
If you care, it is life affirmating.
It ADDS to the equasion.

The very radical idea of "free software" or "open source"
has basically to do with this.

Yes, it takes some inner strength to be able to afford to give
something to others without getting a penny in return.
It kinda contradicts the whole "modern world" ideology.

But...

It is not such a "stoopid" idea, just as open source approach
shows. It is alive. It does produce value. And millions of people
enjoy that value and enjoy that very feeling of not being tied
up with their hands to some mega-sucking enterprise, such as
microsoft, google or poogle.

It IS a life affirmative approach.



I bet it would be more beneficial to reconsider the pricing issue.

Sure, free trial gives you at least something.
You don't have to shell out an arm and a leg just to see if
you like something enough to get married to it for a while.

But at the same time, one way or the other, you need to pay for
things. You need to pay for water. You need to pay for electricity.
You need to pay for food.

In today's world, software is like water or food.
And you can not expect to get something for nothing.
I would rather see Sun charging a small fee for Java
instead of giving it for free and eventually going out of business,
being bought by some sharks of Oracle, Microsoft of Google type.

You don't have to go to extremes.
Either you charge an arm and a leg or give it for free.

In an ideal world you wouldn't have to go to extremes, but
this is not an ideal world. Given the pressures that
politicians face from their people to do something to help
them out of their difficult circumstances, the politicians
of countries like China and Russia will throw even large
companies like Microsoft under the bus when it comes to
issues of software piracy. The on line, free model works
out best I think since control over the software is
maintained by the authors and because its free it spreads
more quickly. It is exactly what you want from a business
model perspective. Garage sales are popular where I'm
from. This is 100 times better than a garage sale because
everything is free. People need quality software that
is free just like they need decent but cheap stuff at
garage sales.


Brian Wood
http://webEbenezer.net
651) 251-9384

 

[Brian Wood]

I never said they were the majority.  But everything on my home and work
desktops and servers is open source (OpenSolaris and Linux) and quality
wise it's up with any closed source alternative.  

I'm using Linux -- Fedora 12 -- but I don't think it is very
good. It hangs on me fairly often -- it seems to have a
problem with extensive use of ssh -- and I have to reboot.
I guess OpenSolaris is better, but Sun is going out of
business and has to be bought out.

The products my
company ships are based on a plethora of opensource packages and build
with gcc plus the one large piece of smelly proprietary software.  We
have the source for the latter, so we know how bad it can be!

Gcc isn't much to write home about.

Based on my fairly extensive experience,

This reminds me of David and Goliath. Remember Goliath was
the experienced warrior. David was the young man who didn't
want Saul's armor because it slowed him down too much. He
would have been a sitting duck in that.

the organisation of open source
projects and commercial firms is pretty similar.  There is a similar mix
of chaos and order.  It comes down the the individual(s) who drive the
projects.

I don't like the phrase "drive the projects." That sounds like
the tyrannical Apple again. Whether it's politically or business
wise, kind/thoughtful leadership is hard to find. "The Bad Lady
Flies a Boeing 757" is another example of poor leadership that
we've had to endure in the US.


Brian Wood
http://webEbenezer.net
(651) 251-9384