[SOLVED] (4?) Things that support the Four Freedoms

I don't personally think it's any better to be adding to the "4 freedoms" dictating how software is written. It's a very slippery slope that actually detracts from people's freedoms. And it isn't any better than M$ dictating how you can and can't use their software, and/or dictating "standards" to railroad people into doing things "their way", that only benefit them.

At the end of the day, if the demand is there, then there's still reason for GNU developers to continue developing GNU software. And therefore, nobody is holding a gun to anyone's head - if you don't like something, don't use it. Use something else that does fit your criteria, and/or get involved in the GNU project and write some code. There's little point in whingeing about the "4 freedoms" not being enough - the people that make the decisions are hardly going to care what one person thinks. Let alone the likes of IBM, Red Hat, etc - if they did care, they'd already be doing everything the way you "approve of".

It never ceases to amaze me how much people want to whinge and complain about something, but yet don't want to actually take any meaningful action to provide a viable solution instead of just whinging and complaining about it - it's absolutely pointless and changes nothing.

Originally Posted by jsbjsb001 I don't personally think it's any better to be adding to the "4 freedoms" dictating how software is written. It's a very slippery slope that actually detracts from people's freedoms.

I agree.

And it isn't any better than M$ dictating how you can and can't use their software, and/or dictating "standards" to railroad people into doing things "their way", that only benefit them.

You do realise, that the reason this is being talked about is because of corporations strategically co-opting software development to shut out existing developers, right? So... you're saying being against that is um, no better? Hmm...

if you don't like something, don't use it. Use something else that does fit your criteria

Hobson's choice. Nice. So before we had freedom, and now we have "take or leave it." I'm glad you're not in charge of this.

The most cynical thing about saying "Use something else" is that the very problem we are talking about (lack of modularity) can shut out the very solution youre proposing. It's the software equivalent of telling starving people who have no bread "let them eat cake." And that kind of attitude is going to catch up with these corporate developers eventually.

It never ceases to amaze me how much people want to whinge and complain about something, but yet don't want to actually take any meaningful action to provide a viable solution

Trying to come up with a meaningful solution is exactly what's being attempted here. The closest anyone is coming to whinging and complaining here is...

At any rate, I've got the answers that I was looking for. I don't think [SOLVED] is appropriate, because this problem will persist for some time. But at least the thread was useful to the OP, which is me.

Sorry if you take that as whinging. But not much.

Originally Posted by freemedia2018 They aren't numbered 0 to 3 just to be computer-geeky, as many assume. Freedom 0 is 0 because they considered it so fundamental.

And yet so fundamental that they initially forget about it...?

The current page claims:

Around 1990 there were three freedoms, numbered 1, 2 and 3. Then we realized that the freedom to run the program needed to be mentioned explicitly. It was clearly more basic than the other three, so it properly should precede them. Rather than renumber the others, we made it freedom 0.

And yet this does not agree with the Wayback Machine, which shows three unnumbered freedoms as late as February 1999, with the new freedom being present in April 1999 (along with explicit numbering):
https://web.archive.org/web/19990224...y/free-sw.html
https://web.archive.org/web/19990430...y/free-sw.html

*shrug*


Anyway, I have not seen a compelling reason to not add a fifth freedom, nor to avoid reformulating the existing ones.

The current page opens with: "...users have the freedom to run, copy, distribute, study, change and improve the software"

Why are copy and distribute spread over two freedoms (when they can be expressed as one), whilst study and change are squashed into one when they are independently important?

Here would be my fifth: replace - the freedom to not run the software, to be free to avoid vendor lock-in through appropriate modularization/encapsulation and minimized dependencies - meaning any free software can be replaced with a user's preferred alternatives.

Originally Posted by boughtonp Here would be my fifth: replace - the freedom to not run the software, to be free to avoid vendor lock-in through appropriate modularization/encapsulation and minimized dependencies - meaning any free software can be replaced with a user's preferred alternatives.

I've been searching for an answer that good. For months. I knew it could be simple, I just didn't know what on earth it could be. You might want to tell me what kind of attribution you want for that, until then it's "boughtonp."

Isn't the freedom to replace dependent on modularity?

Originally Posted by freemedia2018 You might want to tell me what kind of attribution you want for that, until then it's "boughtonp."

That's fine, or if the context prefers a name then "Peter Boughton" also identifies me.

But no need to jump and engrave anything in stone straight away - it might be a good way of expressing what you've been trying to say, but that doesn't mean it can't be improved.

Originally Posted by hazel Isn't the freedom to replace dependent on modularity?

Perhaps, but I'm not sure what you're trying to say.

Modularity itself is not a user freedom, it's a structure/property, and it can mean different things to different people.
A system can be segmented and described as modular whilst still being inter-dependent/over-coupled and difficult to switch out.

Describing user freedoms should start with the goal, before extrapolating on the prerequisites/methodologies needed to achieve it.

If you look at my earlier post, you will see that I define modularity as meaning that user-level packages should not have unnecessary interdependencies. Programs should be dependent only on their libraries, not on having other user-level packages installed. Now I suggested this initially because it makes a system easier to understand. In other words modularity -> simplicity. But it seems that also modularity -> freedom not to use particular software.

Ah, yeah - in that sense modularity encourages simplicity (which helps the freedom to study) as well as helping to avoid being forced to follow other people's preferences.

Originally Posted by freemedia2018 FILO "Free In License Only" or what some people (not me) call OSPS "Open Source Proprietary Software"

I've posted about this here and elsewhere over the years. I have termed it "proprietary in all but licence".

From my point of view it's code which is released under a free licence, but every other aspect of it's development and distribution is proprietary in nature. For example:

1) It is developed for a specific business need and not designed around modularity, portability or the best interests of the end user.

2) It may use a permissive licence as a "bridge" between GPL and proprietary code. Indeed it may be engineered around this in particular, with performance, modularity, portability, not to mention security being less important.

3) It's a "viral" piece of software which has been engineered to be difficult to avoid, with the full cooperation of other pieces of popular corporate backed software - from those with the same/similar agendas and vested interests.

4) As a device driver kernel module, it may also be an implementation which uses proprietary device firmware (also with same GPL-> permissive -> proprietary linkage). This is a classic example where "free and open source drivers" are actually not free at all.

5) Despite its licence(s), it may still bundle adware/spyware/data mining/tracking functionality, which is "opt out", as with many proprietary solutions and geared towards exploiting any laziness or ineptitude of end users.

6) It has been developed as a "product" with the target market being a "consumer" and has deployed Windows registry "hive" style binary configuration and removed many configuration options to ensure that consumer uses the software exactly how the developer wants them to use it and in no other way. Its priority is to push the "brand" first and foremost.

7) It has been engineered to discourage building it from source, understanding it and forking it.

8) It is a software "project" which is "marketed", via lots of weasel words, around the principles of berating and attacking that which is seeks to obsolete/replace.

9) It is "Linux proprietary" and it's developers couldn't care less if it doesn't work on other UNIX-like OS, because they only consider the OS used by their employer/corporate backers to be relevant. For many Linux users, this invokes a shrug of the shoulders, but the reality is that it's not only a matter of being "Linux proprietary", it's a matter of being proprietary to just one kind of Linux - that's the other end of the slippery slope.

10) It is free in licence, but not in spirit.

Originally Posted by hazel If you look at my earlier post, you will see that I define modularity as meaning that user-level packages should not have unnecessary interdependencies. Programs should be dependent only on their libraries, not on having other user-level packages installed. Now I suggested this initially because it makes a system easier to understand. In other words modularity -> simplicity. But it seems that also modularity -> freedom not to use particular software.

One of the things that ive been saying is that it would be ideal to find a quality that "generates" modularity as a function of when it is necessary.

I was already aware that it likely isn't possible, or even desirable, to have perfect modularity in every single instance of everything. I use tinycore as an example. When tinycore first came out, I had just purchased the DSL book (paper and cd) and emailed Robert Shingledecker, who told me his problems with DSL and told me about his then-new tinycore. I've used it many times, I was very excited. But the way that tinycore does modularity is almost tedious. Cool, but tedious.

Between that and the pie-in-the-sky near-impracticality of the HURD kernel (it's really hard to make a successful microkernel-- I'm no Linus fan but I think he had a point there) it leads me to believe that A. modularity can be as tedious and problematic as monolithic design in some instances and B. what we really want is modularity as often as it is beneficial, which is most of the time! Certainly before Init becomes a nasty, "mandatory" GNU replacement.

I was looking for a quality that generated that "modularity when necessary" aspect, when Peter Boughton goes and says "how about the freedom to replace?" Sure, once it's right in front of you, it seems like it was obvious all along. But it's worded really well. And people are already asking "isn't that covered by the freedom to change?" Apparently Not! Because when you talk about how systemd threatens the freedom to change, people say there is no such thing (in that context!)

As was pointed out, the freedom to share copies and share modified copies are distinct for some odd reason. So the freedom to change and the freedom to replace are distinct. But in this day and age-- they really need to be! Because the freedom to replace, to NOT run the software, was always implicit-- and that's gotten us five years of Hell!

This fifth freedom isn't perfect (none of them are IMO) though it's awfully close. So perhaps it will get tweaked slightly. But I'm content. By the way, Denis Roio (Devuan leader, Dyne.org creator for anybody who doesn't already know) loves it. As with the Pillars, if this doesn't make the Four Freedoms into Five, then I'm probably still going to use it as an addition. "Yes, we believe in the Four Freedoms-- and the Fifth! The freedom NOT to run the software." Funny how freedom 0 left that out. (Nice find on the Internet Archive as well.)

Originally Posted by freemedia2018 This fifth freedom isn't perfect (none of them are IMO) though it's awfully close. So perhaps it will get tweaked slightly. But I'm content. By the way, Denis Roio (Devuan leader, Dyne.org creator for anybody who doesn't already know) loves it.

Ah well, then no reason to look further. Name dropping usually does not add credence the speaker/writer was hoping for.

Originally Posted by cynwulf I've posted about this here and elsewhere over the years. I have termed it "proprietary in all but licence". From my point of view it's code which is released under a free licence, but every other aspect of it's development and distribution is proprietary in nature. For example: 1) It is developed for a specific business need and not designed around modularity, portability or the best interests of the end user. 2) It may use a permissive licence as a "bridge" between GPL and proprietary code. Indeed it may be engineered around this in particular, with performance, modularity, portability, not to mention security being less important. 3) It's a "viral" piece of software which has been engineered to be difficult to avoid, with the full cooperation of other pieces of popular corporate backed software - from those with the same/similar agendas and vested interests. 4) As a device driver kernel module, it may also be an implementation which uses proprietary device firmware (also with same GPL-> permissive -> proprietary linkage). This is a classic example where "free and open source drivers" are actually not free at all. 5) Despite its licence(s), it may still bundle adware/spyware/data mining/tracking functionality, which is "opt out", as with many proprietary solutions and geared towards exploiting any laziness or ineptitude of end users. 6) It has been developed as a "product" with the target market being a "consumer" and has deployed Windows registry "hive" style binary configuration and removed many configuration options to ensure that consumer uses the software exactly how the developer wants them to use it and in no other way. Its priority is to push the "brand" first and foremost. 7) It has been engineered to discourage building it from source, understanding it and forking it. 8) It is a software "project" which is "marketed", via lots of weasel words, around the principles of berating and attacking that which is seeks to obsolete/replace. 9) It is "Linux proprietary" and it's developers couldn't care less if it doesn't work on other UNIX-like OS, because they only consider the OS used by their employer/corporate backers to be relevant. For many Linux users, this invokes a shrug of the shoulders, but the reality is that it's not only a matter of being "Linux proprietary", it's a matter of being proprietary to just one kind of Linux - that's the other end of the slippery slope. 10) It is free in licence, but not in spirit.

outstanding!

Originally Posted by ChuangTzu Ah well, then no reason to look further. Name dropping usually does not add credence the speaker/writer was hoping for.

I don't know what you're getting at, CT, Stallman is never going to approve a Fifth freedom, I don't trust anybody left at the FSF (and they won't change it either) and Roio's approval is a positive IMO.

Why turn that into an accusation of name-dropping? The reason wasn't anything as petty as that. The actual purpose here is to try to solve a problem and Roio is a free software philosopher, not completely unlike Stallman. Is it even possible for his opinion to be relevant without you implying such petty intentions? Relaying a message from someone often involves their name. Should I have said that "a random person I won't name liked it too?" Or do you really think Roio's opinion has no relevance? I certainly don't agree. I'm disappointed that you felt a need to imply what you did.

Originally Posted by freemedia2018 I don't know what you're getting at, CT, Stallman is never going to approve a Fifth freedom, I don't trust anybody left at the FSF (and they won't change it either) and Roio's approval is a positive IMO. Why turn that into an accusation of name-dropping? The reason wasn't anything as petty as that. The actual purpose here is to try to solve a problem and Roio is a free software philosopher, not completely unlike Stallman. Is it even possible for his opinion to be relevant without you implying such petty intentions? Relaying a message from someone often involves their name. Should I have said that "a random person I won't name liked it too?" Or do you really think Roio's opinion has no relevance? I certainly don't agree. I'm disappointed that you felt a need to imply what you did.

Meh, who already agrees does not matter. Good ideas stand on their own merit and I think you have a good idea with adding a "Freedom" but that shouldn't matter to you.