Beefy Boxes and Bandwidth Generously Provided by pair Networks
Perl Monk, Perl Meditation

Re: Copyright on languages

by sundialsvc4 (Abbot)
on May 02, 2012 at 18:50 UTC ( #968508=note: print w/replies, xml ) Need Help??

in reply to Copyright on languages

I think that this is a case where some kind of intellectual property protection may be needed ... because after all it is true that “a lot of hard creative work” goes into the initial design of a thing, whereas subsequently duplicating the established design is trivial.   (All I need to duplicate a roadway is a paving machine, so to speak, and I didn’t have to select and survey the route or clear the underbrush, as you did, in order to now freeload upon your entire effort.   That’s not fair.)   But I don’t rightly know whether Copyright law, much less Patent law, is “it.”   I am not sure that the appropriate body of law has been designed yet.

Replies are listed 'Best First'.
Re^2: Copyright on languages
by ikegami (Pope) on May 02, 2012 at 19:09 UTC

    I think that this is a case where some kind of intellectual property protection may be needed

    Do you think we'd have more or better programming languages today if would-be authors knew their languages and APIs would be covered by Copyright? I think we'd be worse off.

      No, I don’t think that Copyright is necessarily the right form of IP protection, and we have definitely lost control of Patents with regards to software.   (Only the lawyers and vultures are making money.)   Which is why I said that I’m not sure that the necessary IP protection concept has been architected yet.

      The problem as I see it ... and it definitely goes way back even to the “clean-room” cloning of the original IBM-PC BIOS ... is that the first designer and developer of anything that is functional, be it a roadway or a piece of software, incurs a lot of costs and takes a lot of risks that “all those who come thereafter” do not have to experience.   You simply build another railroad fifty-five feet to the left.   The original company conceived of the idea, turned it into a practical implementation, cleared the jungle and sold the idea.   All you did was to exactly copy it.   What you copied, and for that matter your motivation for copying it and the entire source of your profit opportunity for having done so, was none of your doing.   That’s not fair.

      On the other hand, we do not want to squelch “the idea of a road,” nor do we want to squelch the building of better ones.   Which is more-or-less what Patents try to do.   Which is why IMHO software patents are categorically worthless:   they are expensive and don’t accomplish their mission.

      Awkwardly pursuing the road-building analogy:   as an incentive for a company to build a superhighway, I could authorize them to collect tolls for the next five years.   I could stipulate that no roadway can be built within one thousand yards of that same route for the same interval, or stipulate that, should one be constructed, they could collect commensurate tolls on it also for the same period of time.   But we simply don’t have any such notions in IP law right now, AFAIK.

      INAL ... thank god ... but basically I think that what ought to be protected for a little while is the business advantage that comes from “having done all that up-front design and build work.”   You would be afforded a finite window of opportunity during which you would be entitled to royalties.   I’d think of this notion as “pragmatic, reasonable, and fair.”   (Until the designing attorneys chopped it to pieces...)

Re^2: Copyright on languages
by tobyink (Abbot) on May 03, 2012 at 07:04 UTC

    I would have expected patent law to be closer, as patents deal with inventions. If I patented a foot-driven can opener, and you started manufacturing something that worked the same, then I could potentially sue you.

    I imagine the reason Oracle are trying their luck under copyright law is that they never applied for a patent for their API (whereas in Berne convention countries, copyright applies to works automatically). Plus in patent law prior art is important, and the design of the Java language (though not much of the API) borrows a lot from C++ and C.

    perl -E'sub Monkey::do{say$_,for@_,do{($monkey=[caller(0)]->[3])=~s{::}{ }and$monkey}}"Monkey say"->Monkey::do'

      In counter to that notion, I offer the observation that patent law as actually applied to software simply has not worked.   We literally have patents covering the idea of a “twirlybird wait symbol” being translucent rather than solid.   A patent was issued to IBM covering the idea of using a “compare double and swap” instruction to update a singly-linked list even though the identical idea is taught in textbooks.   I opine that “patent law simply has not worked” because the idea is a shoe that really does not fit.   And copyright law really does not fit, either.   I think this is a different beast, such that a new legislative design is needed.   (And it won’t be an easy, obvious, or controversy-free design.)

      Oracle basically wants someway of making money out of Java, anything would be OK. But they don't want a situation where they would want someone to be paying for per installation of their software.

      What they basically don't want is charging per server installation on orders of some thousand dollars per installation. That would kill Java.

      They would ideally want some other way like patent/copyright suits, or Manufacturers paying them per mobile device shipped. That would make a better deal for them, Instead of directly making a end user pay for a installation.

Log In?

What's my password?
Create A New User
Node Status?
node history
Node Type: note [id://968508]
[LanX]: I thought MST perl was obvious ;-)
[RonW]: I have so much accumulated in my memory that finding the right item at a given moment often fails

How do I use this? | Other CB clients
Other Users?
Others meditating upon the Monastery: (8)
As of 2017-08-18 20:40 GMT
Find Nodes?
    Voting Booth?
    Who is your favorite scientist and why?

    Results (310 votes). Check out past polls.