trochee: (bithead)
[personal profile] trochee
[note: if you're already a free software geek, this is undoubtedly review. But I figured if I'd written the explanation, then it's more useful if it's more shared.]

A friend -- who's really quite sharp on computer stuff, given no training -- asked two of us in my lab about classification algorithms and decision trees. The other guy in the lab pointed out IND and MatLab and cc'ed me; I pointed out another:

There is also c4.5, from Ross Quinlan, but he has decided to try to make money off the software so it's not Free as in Speech. (but it's still Free as in Beer, and you can download c4.5 from here).
But the friend said, quite rightly:
[[livejournal.com profile] trochee], I never did get this speech / beer distinction. Explain?
This is what I wrote:
ah, I see the problem. Look at the second paragraph of this web page. "free as in [free] speech, not free as in [free] beer" is an attempt to make the distinction between free/libre and free/gratis.

That is, "free" has two senses in English: "liberated" -- presumably from the constraints of encumbrances restricting the redistribution, like copyright -- and "without charge". Some software is free as in "free speech" without being free as in "free beer" -- the GPL, for example, allows redistributors to make it available for the cost of redistribution.

But the two meanings often apply to the same piece of software: I can download the Gnu C Compiler (gcc) for no money (free as in beer) and I can be sure that it won't be developed in a way that I can't contribute further (free as in speech).

Thus c4.5 is free-as-in-beer: no cost, but not free-as-in-speech because there are strong restrictions on the redistribution or modification of source. RQ doesn't want people patching his code, for example, because he is trying to make a buck off c5.0 and if people keep fixing c4.5 then it will become a community resource faster than he can keep up. Which is (if I may digress again) the reason that (1) c4.5 is not free-as-in-speech and (2) the general reason that a lot of big software companies *cough*micro*cough*soft*cough* are very afraid of letting bits of their code become free-as-in-speech.

does that help?

Well, does it?

Date: 2005-01-28 09:49 am (UTC)
From: [identity profile] soliss.livejournal.com
Well, it made sense without the explanation, to me at least. *shrug* Didn't hurt, though.

Date: 2005-01-28 11:41 am (UTC)
From: [identity profile] congogirl.livejournal.com
Ditto. You were supposed to do that math thing we learned in school where you apply the part outside the parentheses to all items in the parentheses, then continue with the equation.

Semirelevant tangent

Date: 2005-01-28 07:38 pm (UTC)
From: [identity profile] fructivore.livejournal.com
The whole distinction would be better expressed using the terms "free" as distinct from "open".

"Free as in speech" software seldom actually represents a form of public discourse in need of special protection (what we usually mean by "free speech") — what its authors mean is that you have access to how it's written.

Generally this access is given without charge, but that's nothing more than the "beer" connotation.

It would be most correct to say that some software is free, and that some free software is open.

The plausibility of open software which isn't free is left as an exercise to the reader.

Profile

trochee: (Default)
trochee

June 2016

S M T W T F S
   1234
567 89 1011
12131415 161718
19202122232425
2627282930  

Most Popular Tags

Style Credit

Expand Cut Tags

No cut tags
Page generated Jan. 19th, 2026 06:55 am
Powered by Dreamwidth Studios