So to kind of clarify on my original idea a bit. (finally found the words)
Essentially, the point of libre software is that anyone can change it to do what they want the program to do, and, if someone doesn't have the know-how, they can ask someone else.
That's the heart of the idea. There should be an organization not-unlike an artists' guild, where people can go to and commission customizations to libre software.
The idea of a standards organization, was more meant as a certain standard expected of all guild members. Projects live mostly off of donations, but they can certified based on their members' respective statuses in the guild and their portfolio of commissioned mods they've accomplished.
Supposedly major companies could commission a major modified fork from a project (like Blender or Gimp) and that there would be a certification mark the commissioning company could bear indicating that they supported as well as respected free software in this way. (i.e. not trying to convince the foundation to sign an NDA, close or obfuscate parts of the fork, or write any kind of exploitive or otherwise disreputable software.) And, if a company dedicates their own staffed developers to the effort, basically the mark would also indicate that those developers weren't disruptive and were respectful.
On 5/31/17, John Luke Gibson eaterjolly@gmail.com wrote:
Neverminding the ridiculous length of that subject line..
I just thought an interesting thought.
First, a little context, (I know how rms feels about blockchains) I was investigating slock.it and thinking to myself "why don't they just make a hardware standard like eoma instead of closing their development and calling it open?" (Like, Pi-Top is [n]ever gonna release those stl files) (I realize that's a loaded 'just' cause it sounds easy, but is one of the most difficult possible)
Then, it dawned on me: Lulzbot doesn't do that.. Wait, Lulzbot exclusively uses open software in their development.. Then *bam* like a boulder (nothing to do with Lulzbot): GPL-violations, improper GUI training, failing to extend using APIs/Addons, failing to bugsmash/'track-issues', failing to participate in mailing-lists and irc, failing to simply fork when development goals conflict, planned esoteric-ism and/or planned obsolescence, failure to secure clientèle data by using fully free systems (when relevant), failure to participate-in and be-aware-of public conversations about the underlining security of said systems (when relevant), failure to disclose supplychain information/identities (when relevant), failure at general transparency.
All of these things traditionally go wrong with not only companies that use open source, but companies in-general.
Then, it truly truly dawned on me, free software needs standards organizations as well.
On Tue, Sep 5, 2017 at 1:18 AM, Jean Flamelle eaterjolly@gmail.com wrote:
Essentially, the point of libre software is that anyone can change it to do what they want the program to do, and, if someone doesn't have the know-how, they can ask someone else.
The point of libre software is the freedom to do those things. Those things you list are the outcome of having that freedom. I'm nitpicking but I think that distinction is important.
That's the heart of the idea. There should be an organization not-unlike an artists' guild, where people can go to and commission customizations to libre software.
I'm not sure 'guild' would be the right word for such a group, due to the negative connotation of the word in some contexts. Writer's guilds and trade guilds, for example, are quite philosophically different from what the FOSS community does. It could be said that they respected neither freedom nor the art in which they were practicing. Traditional and modern guilds have been criticized for being cartel-like and rent-seeking.
The word might have more or less stigma depending on locale. Where I live, it's a pretty derogatory term in that context.
As far as commissioning software goes, I think that a lot of end-users of open-source software that don't necessarily program or contribute much would probably like that idea, if they had some sort of a system to organize those ideas. If it were a website, maybe it would look sort of like Kickstarter and/or Indiegogo, but where people could vote without giving any money, as an option, with possibly a donation button and donation amount tracker indicator, where the user could donate in bitcoin or some other hopefully more convenient manner.
There's work that I would commission today if there were a popular platform to put it on. Certification would be a plus, but honestly, if it's licensed GPL or similar, I'm already more than happy can't can't really ask for much more than that.
Hardware? That's a whole 'nother story. That's what really deserves well- written and well-followed standards. That's where free and/or libre organizations should be focusing their time. Current standards for it are:
1) vague 2) not 100% libre 3) Violated without repercussion
Neil Jansen njansen1@gmail.com writes:
...
As far as commissioning software goes, I think that a lot of end-users of open-source software that don't necessarily program or contribute much would probably like that idea, if they had some sort of a system to organize those ideas. If it were a website, maybe it would look sort of like Kickstarter and/or Indiegogo, but where people could vote without giving any money, as an option, with possibly a donation button and donation amount tracker indicator, where the user could donate in bitcoin or some other hopefully more convenient manner.
Great care needs to be taken when considering paying people to do things that they might otherwise do for the love of it.
If you introduce a monetary incentive, and the work is then done by people who's primary motive is money, then while the volume of contributions might well go up, the quality could plummet.
You may then find that those unpaid heroes that stand as gatekeepers, triaging patches that come into free software projects, suddenly find themselves drowning in excrement.
They may be just a little upset when they discover that the reason their lives just got noticeably worse it that other people are being paid to do that to them. If the problem is sufficiently bad they may lose their motivation.
Net result: the project that you were hoping to help gets some crappy patches, and loses some long-term contributors.
Perhaps my view is overly skewed by they way it didn't work for Debian, with Dunc-Tanc -- see for instance:
https://lists.debian.org/debian-devel-announce/2006/10/msg00026.html
On the other hand, Debian's LTS _is_ done pretty-much in this way, and doesn't cause much angst -- but it's very clear to everyone that nobody was doing that job before someone organised a way to pay for it.
Cheers, Phil.
On 9/5/17, Philip Hands phil@hands.com wrote:
Great care needs to be taken when considering paying people to do things that they might otherwise do for the love of it.
If you introduce a monetary incentive, and the work is then done by people who's primary motive is money, then while the volume of contributions might well go up, the quality could plummet.
You may then find that those unpaid heroes that stand as gatekeepers, triaging patches that come into free software projects, suddenly find themselves drowning in excrement.
Which I think that illustrates the need for more guild-like behavior, where notoriety and respect of skill is proportional to one's influence networking commissioners.
The way to avoid an organization similar to modern guilds is to ensure sincere meritocracy, and the way to do that is complete, uncompromising, as well as loud transparency.
On Tue, Sep 5, 2017 at 6:53 PM, Philip Hands phil@hands.com wrote:
Great care needs to be taken when considering paying people to do things that they might otherwise do for the love of it.
If you introduce a monetary incentive, and the work is then done by people who's primary motive is money, then while the volume of contributions might well go up, the quality could plummet.
can and has. thanks for raising this point, phil. there are a number of projects where this has genuinely happened. the primary one i recall is the KDE4 project, which received a $EUR 10m EU Grant about... 10(?) years ago.
they implemented a look-alike copy of *the* worst and most hated version of Windows that has ever been created: WIndows Vista.
look also at what happens with redhat. the employees of many of the projects dominated by redhat basically rush ahead "coding without thought for the consequences".
it seems to be the [extremely general, brush-sweeping] case that there is a certain limted-per-time-unit amount of "creativity" - i.e. retrospective and reflective thought - that any one individual is capable of.
if you *compress time* in which they *act* on those thoughts - for example by paying them money to work on something - then you are genuinely, genuinely in danger of doing two things:
(1) decreasing the "number of reflective thoughts per unit of actual OUTPUT per hour"
(2) decreasing their energy and capacity to HAVE "reflective thoughts per unit of time"
there are some really famous cases of people doing all the thinking first, followed by doing 100% of typing non-stop, second. these include bram cohen (creator of bittorrent) and dr richard stallman (famously known for overloading, filling and having to wait for the keyboard buffer of the terminal on which he was typing source code, straight from his head without ever having written a single word of it down on paper or keyboard before beginning)
these people are exceptions. for pretty much everyone else, actually receiving money and using it to work *full time* on a project is a disaster.
so my advice would be - if anyone's asking - if you _are_ looking to fund someone, make sure that they're doing it on a non-full-time basis, or at least look for _some_ sort of sign that there is at least a huge amount of thought and planning gone into what they are asking you to back.
l.
arm-netbook@lists.phcomp.co.uk