Question on License system

Anything and everything that's related to OGRE or the wider graphics field that doesn't fit into the other forums.
Post Reply
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Question on License system

Post by Robert Eastwood »

And this can be asked easiest by explaining the situation.

Most of my code from years ago when I coded is elsewhere, and the only Rect class I have is some class I threw together to fix a printer with some issues on output, and when doing that I went back and forth between left top, width height, and left top right bottom, so it is not reusable, only workable for that case.

So when I needed a rect class to move around some dialogs, I figured I would do another rect class.

Then I notice that Ogre uses Templates for its Rect class, and that sounds like a good way to do that, so then I look at common.h, and once I see that, it is obvious that is the best way to do a Rect class. So since I am rusty with templates, I look at the code, since I have to see templates to remember how they work after so many years of not coding.

Here is the problem, the only way to do a Width and Height is also in that code, obviously that would be the way to do it. And I notice he uses some macros for min max, for unions and presumably bounding of two rects. Again that is pretty much the only way to do it, and the same way I have seen done before.

So from seeing the code, I am reminded of the structure of writing an intersection and union of rects, (although I think the code in common.h does not check for every situation, it only checks one side of the no overlap condition) Anyways, the way to do a Rect class, there really is only one way to do it, it is the same way Microsoft does it, same way MFC CRECT, and a few wrappers for RECT over the years also did the same stuff, and probably a few other programs, although it was when seeing the code in common.h that my memory was refreshed on some objects like templates, and usage of the min max macro.

How does that fit into license is the question? If using the best algo learned from reading some code, is the same as using the code, then at what point does the idea of 'obvious best application' or like in patents, where if it is an approach many people use it is not part of some patent.

Do you understand the question? If I look at the code, I will remember how to do things, and some of them will be similar to the code, but that is not using the code, except as a learning and remembering tool for ways things are often done. Although if using some best case algorithm is restricted, then is the license saying that looking at the code, makes any work product similar to the code then under some license claim?

Note this isn't to avoid license, if I use something, and the entire rendering part I will probably use, it will be in a library with copywrite and source included, although where is the line of 'substantial' and the difference in learning an algorithm, and applying the same code.

I think if I can look at the code, and replicate a similar algo by understanding the concept, that is learning from the code, although is that the implied meaning of the license?

I ask this, because after 10 years, and quite a few cooked brain cells, I have to look at code to remember some of the easier stuff, templates, and things like passing an object as a constant reference as a better style then as a pointer, and a bunch of other stuff. But I also see the algos being used when I do that, and I understand the algos, and once I do, then how is that a part of the system of the license.

Or I could ask is looking at the code, the same thing as 'using' the code, when it is looked at to remember procedures or styles? That is another way to ask the question. Since I think the style of the code should match some pattern in use in many places, and code has to be looked at to do that.

Anyways, any comments would be appreciated, to better understand the mechanisms of such systems.
User avatar
c6burns
Beholder
Posts: 1512
Joined: Fri Feb 22, 2013 4:44 am
Location: Deep behind enemy lines
x 138

Re: Question on License system

Post by c6burns »

Dude are you testing out an algorithm that generates conversations, or is this actually how you talk and ask a question? I was almost certain from your other thread you were a robot except that you were able to respond and reference things people said.
Robert Eastwood wrote:I ask this, because after 10 years, and quite a few cooked brain cells
Quite a few ... Ogre is MIT licensed. I don't understand what there is to even ask about :shock:
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

c6burns wrote:Dude are you testing out an algorithm that generates conversations, or is this actually how you talk and ask a question? I was almost certain from your other thread you were a robot except that you were able to respond and reference things people said.
Robert Eastwood wrote:I ask this, because after 10 years, and quite a few cooked brain cells
Quite a few ... Ogre is MIT licensed. I don't understand what there is to even ask about :shock:
I am a person that learned to think procedurally, they called me computer and Spock when I was 11 and 12 years old, and started programming when I was younger then that, so formed much of my thought patterns around the idea that most people use pattern matching and synthesis.

My post in this thread is an attempt to understand the procedural aspects of the licensing system, so that I can apply its conceptual intent to the usage of code under that license in a fair way.

The other thread explains better what you call 'robot' thinking. What you are actually noticing is that robot thought is modeled on human thought that few people understand is how they actually think, but is easier to see in a robot.

You do the same thing, you just don't know nor control your use of pattern matching, weights, and relevance, since it is all done in your subconscious mind before exposed to your conscious mind. If you can access the actions and procedures of the pattern matching and synthesis of the subconscious mind, you can actually begin to program your neural networks that is how you actually think, by some of the same methods you mention as 'human modeled' robotic methods.

The advantage to understanding your pattern matching and methods of combining items and selection of relevant context, allows for finding more ideas in some area, that the filter between the subconscious and conscious mind usually filters out. And also to find where that filter is in error when it filters by a 'strong pattern' with 'low relevance' but does not weigh the correlation correctly by the pattern being more important then the actual weight of the relevance of the pattern.

Some examples using real world thought patterns, some person said in an interview, "why would they have an affair, they have a loving relationship in there marriage" What that person is doing, is presuming that her action of infidelity is based on a 'weighted goal' without understanding the actual reason. In that case the individual presumed the 'social' weight of what 'life should be' 'having a stable loving home', was the actual 'goal system' used in her subconscious mind. In her mind the 'goal' was excitement and danger, not a loving relationship, so she did not know that her presumed action was based on her subconscious goals. And instead substituted what she thought her goals were based on what she has been told is how she thinks, not actually how she thinks.

Another example, if you gain joy from 'being top level' in a game, but do it by power leveling, you are taking the 'recorded prestige' that a 'level accomplishes' and that came from the idea of having to 'earn that' and spending that based on 'having the level' not the actual accomplishment that is required to achieve the level. And that 'spends' the conceptual idea of 'level' being an accomplishment, since its actual value of being accomplishment was 'loaded' by many people having to spend much work to achieve it.

So each concept can also be 'loaded' and have 'value spent' based on its perceived value and usage by many people, also a factor of language. If patriot is a group of people that valiantly defended some core ideas, then you slap that label onto a piece of legislation against those ideas. You are 'spending' the contributions of that group, against there actual goals. The original Patriots loaded that noun with value by there virtuous action, then some others try to spend that value, without actually having any correlation to the same core idea.

Again a form of 'value' attached to various nouns in language, and mostly processed subconsciously.

I am actually a person, although the question is are you a chat program?
And could it be we all are that same model, (since the programs try to emulate human thought) but not everyone knows it.

What you are noticing in my posts is a level of correlation that is wider then what your filter between your subconscious and conscious allows you to see, so to you it looks like an error(correlation without weight of relevance), It is what is usually, by most filters in peoples minds, considered to be less relevant, but when used to understand core interfaces of different concepts actually does have much importance, and can show the deeper core interfaces into conceptual ideas behind different stories. (deep core drilling - Armageddon) Such actions are referenced as the guy holding up the fuzzy fingers in Patch Adams movie.

The only reason I reference many stories, is because I have them all remembered by the various different multiple concepts each scene represents(as I see it). And compare and group them by that, not by the linear flow of some movie, or its space or time location.
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

c6burns wrote:Dude are you testing out an algorithm that generates conversations, or is this actually how you talk and ask a question? I was almost certain from your other thread you were a robot except that you were able to respond and reference things people said.
Robert Eastwood wrote:I ask this, because after 10 years, and quite a few cooked brain cells
Quite a few ... Ogre is MIT licensed. I don't understand what there is to even ask about :shock:


The above copyright notice and this permission notice shall be included in all copies or substantial portions of the Software.

What is the definition of 'copy' and 'substantial'.

Does copy include the concept behind the code, such as an algorithm, or only the actual implementation of that algo. If it includes the algo, then the question of is it obvious, or unique enters the equasion. If it is only implementation, then it is to make sure someone understands the algo, then they can use it.

Apocalypse now director Stone, quote to Dennis Hopper, "first you must learn the lines, before you can ad lib the scene." <- same concept, if you understand the algo, you can write it many ways, is it the algo or the 'choice of implementation' that is the part under the license.

That is an easier way to ask the question.
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

c6burns wrote:Dude are you testing out an algorithm that generates conversations, or is this actually how you talk and ask a question? I was almost certain from your other thread you were a robot except that you were able to respond and reference things people said.
Robert Eastwood wrote:I ask this, because after 10 years, and quite a few cooked brain cells
Quite a few ... Ogre is MIT licensed. I don't understand what there is to even ask about :shock:

And honestly, to explain it the explanation also touches on areas of magic.

For example, if actual individuals that are Patriots store value into that 'label'. The noun being a Macro to invoke the idea of being a Patriot in peoples minds, then the word Patriot should have guards that govern its usage, to be sure the label is used in accordance with the base concepts it was built on. That way a different concept can not 'spend' the stored value in the macro label 'Patriot'. when the label is used to reference a conceptual idea.

Most of the implementation of such things are often referenced to as magic, and a good reason to sync concepts (Example: Indiana Jones: so that you grab the wooden cup not the gold cup when looking for the holy grail, however to do that, you must understand the 'concepts' of the 'source of those ideas' and be in the same conceptual ideas). If you do not sync concepts, that sets off traps in the Wizards Keep. Also explained in many stories. To access an interface of some conceptual idea, you must have the same conceptual understanding and purpose, or you will hit a trap. Same thing with Stonehenge, Roman Ideas, Inca Empire, Time Lore, Balance, Curly/Medusa, and many other conceptual ideas represented by various individuals and concepts in society.

The key idea of traps in the Wizard's keep is to keep people from using something meant for one purpose, to instead be used for another purpose. If the purpose is against those that invested time in the creation of the interfaces to the concepts of an idea, traps are triggered. And a 'noun' is a form of interface to the feelings loaded into a concept. Therefor usage of nouns can invoke a responsibility to how they are used.

Hence one of my procedures is to learn the rule sets of an area.
Hence I ask for clarification so I can be in sync with the same ideas as the license intends to support, before usage of some item.
dbtx
Halfling
Posts: 66
Joined: Tue Nov 01, 2011 9:07 pm
x 10

Re: Question on License system

Post by dbtx »

A few questions:
1) How would a non-native English speaker with Asperger's write?
2) Is there a Reverse Turing Test that would-be Turing Test judges have to pass? because I'm failing
3) I hate to be a dirty species-ist, but... Robert, have you any first-hand experience with glucose?
User avatar
c6burns
Beholder
Posts: 1512
Joined: Fri Feb 22, 2013 4:44 am
Location: Deep behind enemy lines
x 138

Re: Question on License system

Post by c6burns »

Honestly, about the original question of licensing... MIT is meant to be easily understood, and it is very permissive. It boils down to "do literally anything you want with this code, including use it as part of a proprietary work, just don't remove the MIT license. Also, don't sue me, the copyright holder!". If you have additional questions beyond that, perhaps you need to contact a lawyer. They will be happy to charge hourly while you explain how people called you Spock at age 11 :D
Robert Eastwood wrote: You do the same thing, you just don't know nor control your use of pattern matching, weights, and relevance, since it is all done in your subconscious mind before exposed to your conscious mind.
Quite to the contrary, I do consciously control my use of pattern matching and relevance. It is what keeps me from spewing 4 pages of irrelevant dialogue when asking a question about licensing on an internet forum. I consciously ask myself "am I providing the most relevant details in order to receive assistance from others?"
Robert Eastwood wrote:What you are noticing in my posts is a level of correlation that is wider then what your filter between your subconscious and conscious allows you to see, so to you it looks like an error(correlation without weight of relevance)
Now you are being condescending? The fact that I consider 99% of what you write to be irrelevant isn't because I can't see the relationships between the concepts about which you are talking. It's because the correlation between the data you present is so weak that it does not require presentation in order to help provide you with an answer to your question. An example of this behaviour would be me asking a question about a rendering concept (like index buffers for example) and somehow including a 10 page story about a childhood experience I had at the age of 5 where I peed my pants at school. That experience may indeed be correlated to my understanding of index buffers, but it is completely irrelevant to everyone around me so I choose to keep it to myself :lol:

Anyway, I hope you find your answer. Sorry I picked on you :)
User avatar
Kojack
OGRE Moderator
OGRE Moderator
Posts: 7157
Joined: Sun Jan 25, 2004 7:35 am
Location: Brisbane, Australia
x 534

Re: Question on License system

Post by Kojack »

Copyright applies to the actual code implementation itself. The concept behind it would be covered by patents (I want to say nobody would patent a rectangle class, but apple or amazon will probably prove me wrong sooner or later).
User avatar
c6burns
Beholder
Posts: 1512
Joined: Fri Feb 22, 2013 4:44 am
Location: Deep behind enemy lines
x 138

Re: Question on License system

Post by c6burns »

If you can show prior art on a rectangle, I'd be interested in reading that patent application.
User avatar
Kojack
OGRE Moderator
OGRE Moderator
Posts: 7157
Joined: Sun Jan 25, 2004 7:35 am
Location: Brisbane, Australia
x 534

Re: Question on License system

Post by Kojack »

Recently Amazon was granted a patent on photographing things against a lit backdrop (one light on the subject, one light behind the subject aimed at the backdrop). Something that photographers have been doing since long before Amazon even existed.

Hmm... I was joking about the rectangle thing, but a quick search came up with this: http://patents.justia.com/patent/6914617
That's a patent on determining if two 2d rectangles overlap by treating one rectangle as a point and growing the other rectangle by the size of the first. I've been telling my students to do that with various collision shapes for years.
The patent was granted in 2005, so it's valid until 2025. Be careful how you test rectangles, you could get sued.

Here's one from 1996 that does a similar thing, it works out if rectangles overlap using just 2 corners of each rectangle (because they claim it normally takes 16 tests to see if 2d rectangles overlap).
http://patents.justia.com/patent/5522020
dbtx
Halfling
Posts: 66
Joined: Tue Nov 01, 2011 9:07 pm
x 10

Re: Question on License system

Post by dbtx »

Sorry, apparently I didn't get to read something in between (was deleted) but I'll take that as a firm 'yes' for question 3, and I'll stop playing.

As far as licenses and patents, doesn't it ultimately depend on too many unknowns (citizenship, prior art, which lawyers are paid the most, whether the judge had enough coffee)? I'm sure this comes up on Stack Exchange often enough... sometimes you're not even allowed to have seen some code if there needs to be proof that it (being protected work) can not have directly inspired another implementation. Other times it boils down to such as how the for() loop in C can't really be unique without also being wrong and useless, and so you can't patent for(). On a loosely similar note, someone once said that [e.g. transportation] companies will rarely attempt to design and integrate a Tesla turbine because the original is so simple-- the only way to get a new patent so that it's not available to the competition is to change or add enough "features" that they ruin it. But I can't vouch for that person's understanding and mine is quite limited... and then, (mentioned Amazon thing but Kojack beat me to it), so a lot of people's understanding might be outdated.
User avatar
c6burns
Beholder
Posts: 1512
Joined: Fri Feb 22, 2013 4:44 am
Location: Deep behind enemy lines
x 138

Re: Question on License system

Post by c6burns »

Patents are infuriating. I don't know a single substantial software company that hasn't worked with a patent lawyer. I can only imagine that field of law will continue to grow, given the inability of patent offices to adapt, especially in the area of software. Even the US Patent Office acknowledges the issues ... I mean you'd have to be blind not to see them. Yet it remains a difficult problem for them to solve, and one I'm glad not to be tasked with solving. Even worse than companies like Amazon / Apple / RIM / etc are patent troll companies who never intend to produce anything of benefit to society ... just litigate against other companies that do.

A while back on the bullet forum, someone wanted their physics paper reviewed and mentioned they were considering patenting some of the techniques which it described. Pretty much everyone refused to read it on the grounds that they didn't want to expose themselves to infringement liabilities :lol:
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

c6burns wrote: Now you are being condescending? The fact that I consider 99% of what you write to be irrelevant isn't because I can't see the relationships between the concepts about which you are talking. It's because the correlation between the data you present is so weak that it does not require presentation in order to help provide you with an answer to your question. An example of this behaviour would be me asking a question about a rendering concept (like index buffers for example) and somehow including a 10 page story about a childhood experience I had at the age of 5 where I peed my pants at school. That experience may indeed be correlated to my understanding of index buffers, but it is completely irrelevant to everyone around me so I choose to keep it to myself :lol:

Anyway, I hope you find your answer. Sorry I picked on you :)
I don't think you picked on me, but nice to make that comment.

I agree the comment about history is only relevant to respond to the process, and not something I usually do. And even in my post, my only comment was to respond the exact comment that someone posted.

"Is that how you 'always' ask some question."

Therefore a history then makes sense. He should not ask that question if he does not want that reply. It seems people are not actually reading others comments, would you expect me not to answer that question? That implies he wants a history, so my comment was an attempt to answer his question in the form he asked it. He could be more concise if he did not want that question answered.

Condescending? In the first 20 replies the sum total of responses have been between spam, bot, and crazy. Although I appreciate you answering the question, within your understanding. As far as lawyers, they would probably site obscure references as precedence when making a conceptual argument on some topic, so how would that be of any help.

As far as relevance, relevance is also dependent on goals and intent. And if all methods using what I also understand you consider relevant have not achieved needed effects, then expanding further into what would normally be considered less relevant would be the logical thing to do.

As far as ass burgers, or any of the other comments, I would not have spent years in taverns if I had any social limiting afflictions, however I currently am not at the taverns, and I can easily step in and out of different ways to think.


And I do understand where others do not see the relevance.
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

Kojack wrote: Hmm... I was joking about the rectangle thing, but a quick search came up with this: http://patents.justia.com/patent/6914617
That's a patent on determining if two 2d rectangles overlap by treating one rectangle as a point and growing the other rectangle by the size of the first.

Here's one from 1996 that does a similar thing, it works out if rectangles overlap using just 2 corners of each rectangle (because they claim it normally takes 16 tests to see if 2d rectangles overlap).
http://patents.justia.com/patent/5522020
Thanks for the correction on my presumption there was one best way to do those tings, the idea of point in rect is pretty good also, but it does not easily return the union, but is pretty good hit detection method, will remember that one.
Robert Eastwood
Gnoblar
Posts: 7
Joined: Wed May 07, 2014 5:26 pm

Re: Question on License system

Post by Robert Eastwood »

dbtx wrote:Sorry, apparently I didn't get to read something in between (was deleted) but I'll take that as a firm 'yes' for question 3, and I'll stop playing.

sometimes you're not even allowed to have seen some code if there needs to be proof that it (being protected work) can not have directly inspired another implementation.
That was the probable reason for the question, I do not know the level of 'future claims' that are the intent of various licenses. Although I agree most of the current laws are being used against many of there intents anyways.

I presume if the case was that 'any code seen' implies' license on any work product', that would have to be explicitly mentioned, or responded on when asked for clarification, so I will presume code can be used to learn ideas from. Then if used as a functional piece, the headers would be included.

As far as deleted, or what you are talking about as question 3 or 'playing' I don't know what you are talking about with any of that.
User avatar
mmixLinus
Silver Sponsor
Silver Sponsor
Posts: 199
Joined: Thu Apr 21, 2011 3:08 pm
Location: Lund, Sweden
x 12
Contact:

Re: Question on License system

Post by mmixLinus »

Robert Eastwood wrote:So from seeing the code, I am reminded of the structure of writing an intersection and union of rects, (although I think the code in common.h does not check for every situation, it only checks one side of the no overlap condition) Anyways, the way to do a Rect class, there really is only one way to do it, it is the same way Microsoft does it, same way MFC CRECT, and a few wrappers for RECT over the years also did the same stuff, and probably a few other programs, although it was when seeing the code in common.h that my memory was refreshed on some objects like templates, and usage of the min max macro.

How does that fit into license is the question? If using the best algo learned from reading some code, is the same as using the code, then at what point does the idea of 'obvious best application' or like in patents, where if it is an approach many people use it is not part of some patent.
As a programmer, I feel these are classic programmer questions regarding source code copying.
  • Learning the best algorithm from reading some code is NOT the same as using the code
  • You may create your own code based on any algorithm as long as the algorithm isn't patented (or if you have an agreement of sorts)
  • When you write code, you may NOT write code that is a verbatim copy of someone elses code, UNLESS there is written statement granting you the right to do so
Last edited by mmixLinus on Mon May 12, 2014 9:17 am, edited 1 time in total.
Powered by Ogre3D:
MMiX.Me 3D - 3D Music Player
Galaxy Navigator 3D - 2 million stars (ESA's Gaia satellite)
YouTube|Facebook
User avatar
mmixLinus
Silver Sponsor
Silver Sponsor
Posts: 199
Joined: Thu Apr 21, 2011 3:08 pm
Location: Lund, Sweden
x 12
Contact:

Re: Question on License system

Post by mmixLinus »

Also, regarding the feasibility of there being a patent in existance for Rect algorithm(s): Regard the first ever recorded public usage of some rectangle methods. Was that usage preceded by a patent application? If no, then "it" isn't patentable any more. Your claim ("Anyways, the way to do a Rect class, there really is only one way to do it") does give a feeling that a Rect class is not patentable due to lack of an inventive step/it not being non-obvious and not Novel.

Not that I have any proof but don't you think the early philosophers (working on GEOMETRY of course!) devised many simple geometry ideas + algorithms + methods, including those that dealt with rectangles..? (I am talking about thousands of years (prior art) here)
Powered by Ogre3D:
MMiX.Me 3D - 3D Music Player
Galaxy Navigator 3D - 2 million stars (ESA's Gaia satellite)
YouTube|Facebook
Post Reply