Did you miss the part where I said I had both personal experiences from living in these areas, AND the easily demonstrable different rates of offenses? What urban area is trying to disenfranchise people? Where are the fundamentalist cults that think a woman should be subservient and silent? Which populations are passing anti-transgender laws despite no evidence of any problem? Which populations banned gay marriage, and which ones produce the people that (more often) refuse to grant licenses for the same?
Racism, sexism, etc are absolutely everywhere, but that doesn't mean they are there at the same rates, and ignoring that won't help us make it any better.
Getting monads into Lisp in a useful way would require the introduction of features that would certainly make it a different language. You may consider this a feature but you'd be arguing with a lot of people who consider it a bug. (I mean that sentence very seriously; no sarcasm, no joke, it is an accurate description of the situation, nor am I implying I would find the idea horrifying... it would simply be a quite significant change to the core type system.)
Are monads an /advance/, or are they just a language feature that might be nice in some cases, pointless or harmful in others -- like any other feature?
Some monads like the maybe monad would be a dramatic improvement over just defaulting everything to nil everywhere. Likewise error monads would nicely complement the existing CL error mechanism.
You can get some monad behavior in CL now. Bind is fairly easy to write in CL but unit is messy. And CL's lack of automatic currying doesn't help.
It's not clear to me that category theory, functional programming, what have you, provide insight into why things work. In fact, it seems like they might cut the opposite way -- moving from vicissitudes to simple abstractions.
Category theory provides the tools (a language, a set of theorems, etc) to understand why certain things are necessarily true, rather than just "true in practice" or true because of a happy coincidence, similar (IMO) to how complexity analysis gives you the tools to know that merge sort necessarily has a certain baseline performance, rather than just happening to run through all your unit tests quickly.
I gotta say that I think this is just not a big deal for most people. I understand that FOSS ideals have appeal, I understand why people would like them to be protected -- but at the end of the day, I do feel like FOSS advocates, as well-meaning as they are, are perhaps investing their concern in the wrong place. FOSS rhetoric is nice, but it just doesn't seem that important.
I'm very biased, since I worked at VMware for nine years.
Nonetheless I disagree with you. I think it's worth actually pursuing this and settling the VMware situation clearly.
Uncertainty is rarely a good thing for anyone in these kind of situations, and so I'm disappointed that there still hasn't been a real ruling on the merits of the case.
Personally I remain convinced that the vmklinux/shim usage pattern is ok by the letter of the license, but I can understand Harald Welte and Christopher Hellwig's viewpoint.
As much as I agree with you, I think it is unlikely to be cleared up to any great extent. In my mind, the main question of whether or not VMware is in violation of the GPL rests pretty squarely on whether or not it is a derived work. That definition, in a legal sense (AFAICT -- I'm not a lawyer) is intentionally vague. Potential violations in this sense will almost certainly have to be tested on a case by case basis. Which is frustrating for everyone.
Like you, I hope for a speedy resolution of the VMware situation, but I'm not so confident that such a resolution will shine much light on similar issues in the future.
So many people take FLOSS for granted- especially in a place like this where most people are just trying to figure out how to get equity in the next pied piper.
The bottom line is that everyone's life is better because of FLOSS. So many of us owe our careers to an army of forgotten programmers who share their time and brain power to build software for the public domain.
If offers artists in other countries an alternative to expensive software like photoshop. It allows a startup to lower the cost of opening a call center by not having to buy Windows licenses. The cancer researchers trying to find a personalized treatment for your colon cancer use FLOSS. Your state government uses software from Microsoft and Oracle. And even those are affected by and improved due to the existence of FLOSS.
I know you wont understand this. But there are a lot of kids reading this too and they should know not everyone agrees with you.
No, I benefit from free software every day. I'm glad that there's software that isn't encumbered by expensive licensing schemes. What I'm taking issue with is free software orthodoxy, or with the notion that there is some sort of "ethical" issue at play with software that is different from the ethical issues at play with the usage and distribution of /any/ kind of product. "Software freedom" is an odd end-unto-itself.
I find this stance odd to say the least. I'd be happy to throw out copyright (and copyleft along with it) -- but to only enforce classical copyright, and not enforce copyleft? That's crazy.
Are you also advocating ignoring OS X license terms (ie: do you advocate running a bought copy of OS X in a vm on non-apple hardware to build iOS/OS X applications)? Pirate Microsoft Windows, SQL Server - or run Oracle databases without paying a license fee?
If not, I can't see how you can support not following the rather simple copyleft licenses. If a company don't want to leverage all the years of developer effort that is given away for free with the Linux kernel - they're free to invest the millions of dollars worth of work to write their own kernel. I mean, if a grad student could write the Linux kernel in the early 90s, surely VMware can write an OS kernel today?
> I gotta say that I think this is just not a big deal for most people.
I'm assuming by "people" you mean pirates like VMWare or mindless end users like yourself, since it's a huge deal for developers. When I make the conscious decision to license my work for others to use under the GPL instead of something completely permissive like MIT, I clearly don't want some commercial enterprise ripping off my work and using it to line their own pockets without even contributing back their modifications.
I agree that that is irksome. That being said, we're all victimized by commercial enterprises constantly -- it's difficult to see /this/ particular violation as uniquely worthy of outrage (although I confess that this is not a particularly good argument that FOSS violations are uniquely /un/worthy of outrage).
Abiding by licenses is a big deal, and as far as the article is concerned VMWare haven't.
I'm anti-GPL, anti-RMS, but find it as obnoxious as anyone else when a company not only goes with breaking the GPL, but admits it and discards all civility ("take us to court").
inb4 the Stallman disciples descend upon this comment.
The things with FOSS is there are different degrees of rigidity. On the one hand, you have copyleft licenses like GPL that tend to make commercial development very tricky. On the other hand you have more permissive licenses like MIT and Apache that do encourage further open source work but where the vast majority of improvements don't get committed back upstream (since they become a part of proprietary offerings).
That's the difference between OSS and Free software. One focuses just on the benefits of working openly, the other is about making sure the user have full control of what their machine does. Open Source is a side effect of Free and it's a selling point to creators but the consumer is the priority.
Edit: To be clear I'm talking more about philosophy, not licensing. I do still struggle with the idea of GPL requiring changes to be made public.
GPLv2 does not require you to make your source changes public, it only demands that this remains an option for anybody who acquires a license from you.
So if I sell you software under the GPL, I have to provide you the source. You are then allowed, but not required, to publish the source.
Per-copy licensing, which is the base of most commercial software distribution, is pretty much impossible with the GPL. Also, a software company might have licensed some parts of the code from another party - this might not be compatible with GPL. Finally, some companies like to keep trade secrets. This is also incompatible with the GPL.
I can understand that many here disagree with this, and I am an open source advocate myself. Nevertheless, it is important to understand how some companies work. The best way to change this, is to create enough open source software, which can be used as a replacement for closed packages.
> Per-copy licensing, which is the base of most commercial software distribution, is pretty much impossible with the GPL.
Equivalent basis for support contracts, which have actually been the key part for many years in most non-consumer software contracts even when they nominally include copyright licenses, are quite possible (and common) with GPL software.
When you are selling a support contract, GPL is no problem. But this does not cover the sale of a software which costs many thousands per seat, you cannot make up lost sales revenue with support contracts. The cost pressure would drive customers to use the software with way more seats than they have support contracts for. And the GPL would prevent any legal licensing scheme which would try to enforce a support contract per user of the software. There are exceptions to this - a hosted platform like github can of course count the number of accounts created for invoicing.
You're not paying for the software on that server.
You're paying for a license to use a few non-free packages containing trademarked material (basically the visual skins / logos nicely factored out) but more importantly, support.
And the success of Red Hat comes to no small amount from other software companies, which certify their software products often on Red Hat only. That means, if you are trying to get technical support for that software, it might depend on running on red hat. Of course, the software would in principle run on most other Linux distributions, especially Fedora, but I have seen smaller incompatibilities when running not on Red Hat. So this creates a certain market pressure to use Red Hat.
RedHat is smarter than everyone. The have obtained security accreditation that is required by the USG and many commercial auditors. Ultimate lock in and profit from a protected position. How many RHEL license are bought by the USG and the Fortune 500 for only this reason?
RH, the compamy, is a good citizen. These rules only become onerous if you have money, so it is accepted.
Not really. Customers are more than free to switch to CentOS, or even other distributions or pay someone to fork parts of Red Hat.
Red Hat is providing support.
If you modify some free software, repackage and resell it without contributing the changes, then you are breaking the intentional viral effects of the license.
Most of the web stuff that people work on these days is rather outside of this problem, unless is an AGPL license.
It is tricky if you want to base your commercial product on a derivation of a GPL licensed software.
Not really, GPL only says whoever receives binaries, must also receive their source code. Receiver then can, but is not obliged to, publish it or contribute back. Still, yes it does cause problem with various licensing schemes.
But per-copy licensing without source code availability is an atrocity enabled only by misapplication of copyright/IP laws, where copyright holder can write practically anything into the EULA and have the law system enforce the terms "for free". I hope our civilization eventually comes up with better and more GPL-aligned software distribution model.
You don't have to contribute GPL changes back. Whomever you give a binary to you'll have to give the source to if they ask. It's pretty much that simple.
That's a distinction without a difference as anyone who receives the binary must also have access to the source and be free to redistribute that source, so effectively all derivative works of GPL'd code are GPL'd unless you refuse to distribute.
Could you elaborate? What specific concerns you disagree with? What are the importation issues that, in your opinion, FOSS advocates fail to take into account?
