Richard Stallman is responsible for the shrinking economy

Yes source licenses for the compiler itself.

Regards,

-- Walter Banks Byte Craft Limited

Well, that's what I thought.

Are you claiming these - universally detested - measures are somehow of some benefit to a user? I think you are just being obtuse here.

--

John Devereux

Because they often turn out to be a hassle/annoyance for legitimate users. Illegitimate users won't care because they will use a hacked version with whatever protection scheme being disabled (and to that one doesn't need to have access to the source code).

BTW. I don't mind paying for software and I do understand why companies try to protect their software, it is just that many protection schemes tend to hurt legitimate users more than illegitimate users.

Evidence please. I have lots of customer who do not "universally detest these"

No they protect the licensee just as the restrictive GPL license does.

Or are you advocating piracy and theft?

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

I said "usual", and part of this thread was about tools, i.e. compilers. What compiler includes source? In reality, toolchains sometimes even include stuff which the compiler maker is not even allowed to redistribute. The test-suite is one such case.

That aside, of course you can buy all those privileges for money. A lot of money sometimes. With GPL'd tools, you already have the privileges.

Then let it be "Restricted evaluation product". I was using it as an established technical term (and I'm not a native English speaker, so I may not always get all nuances right).

If you have source, and all permissions to do with it what you please, it surely survives the maker's bancrupcy. If the company goes down taking its license servers with it, the product is gone.

Then enlighten me.

The software I do in my spare time (a PBEM game, and some experimental compiler stuff) has no overlap with the software I do at work (file systems and audio codecs).

Bullshit. What I do in paid-for time belongs to my employer, what I do in my spare time is my business - as long as it's a different field of work.

Of course it's unavoidable that I transfer knowledge from here to there. But I take care not to transfer code in the wrong direction. Whereas my hobbyist code sits nicely on my website for download from a work PC, there's no way I could download my work repository from home. Simple. And knowledge? Not more than a neat C++ trick discovered here or there. Nothing big enough to be covered by any IP law.

What "risk"? People who don't release the source code to their GPL-based products simply violate the license. That's essentially the same as people using a commercial product and violating the license, by using it on too many seats, in a product not allowed by the license, on a chip not allowed by the license, etc. You don't talk about a "risk not using your commercial s/w anymore" in this case. You sue them right-away.

By the way, when I send this posting, it will pass through a consumer router based upon Linux. I've got the source tarball somewhere, and the device came with a densely-printed A4 sheet containing GPL, LGPL, and several BSDLs. Why not? And the maker is still in business, doing quite well.

Stefan

Not usually.

However it if is unprotected the vendor looses lots of sales. We can see that in the vast number of people using modified Open Source they do not release.

I have no evidence of this. Do you have any? There is evidence to the country but as you made the statement please prove it. .

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

You do know the difference between having the source code available, and having *usable* source code, don't you? Certainly there is some commercial software (RTOS's and libraries are common examples) where you can make some use of the source code. But unless the license says you can take that code, modify it, re-compile it, and use it in your systems, then the "available source code" is at best a sort of documentation. I know there are commercial packages that have such licensing, but there is a lot (such as Microsoft's "shared source" licenses) which does not.

And do you know that free-price closed source software is basically in the same category as any other commercial closed source software, and bears no relationship to free (as in "speech") or open source software? Actually, I know that you *do* know this - you are trolling here.

Yes.

Yes.

Well most that I know of let you modify it and recompile it.

OK so one does not , many do. Your point is?

No I did not know that YOU put it in the same category. Others do not. Or are you trolling here?

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Lost dongles, broken dongles, dongle driver producing a BSOD when dongle is not present when the PC is started, software not working after hard disk/PC is replaced because software is tied to a particular PC, license not freed from the license server after software crashes, inconvenience of having a dongle sticking out of the laptop when it is moved around...etc; I have seen it all. Yes, those incidents can be solved, often with the help of the vendor. But it does mean lost time, productivity and thus money.

How much sales is lost if a protection scheme is omitted is the question.

First of all a skilled hacker can remove most protection schemes in a few hours and sometimes significantly less than that. Most protection schemes don't offer that much protection against people who intent to use the software without paying for it.

Secondly it is questionable if people who would use a pirated version without paying for it would be willing/able to pay for the software if they didn't have access to a pirated version. More likely they would be more inclined to use a cheaper or 'free' alternative.

Thirdly any serious business would pay for the product if it is worthwhile even if they have access to a pirated version of the software. Not for moral reasons but simply because since there are also other considerations than the tool itself (e.g. support).

People who use a hacked version won't be bothered by issues I mentioned above.

I'm not a FOSS advocate (though I don't mind using FOSS) and I have developed dongle protected software myself, but I do think the criticism w.r.t. protection schemes is fair, as are the reasons for using them.

Quite a few if you want that license.

True of all systems including some Open Source under GPL3 I believe

The test suite is usually third party and something GCC does not have anyway .

And the attendant restrictions

It is not an established technical term it is a derogatory term used by FOSS people to describe commercial eval packages.

Fair enough. I will bear that in mind, I have some experience of German English having worked in Munich

Not so. Recenty a member of this group had a problem with an 18 year old compiler and a dongle. The supplier has long gone. It was sorted out with no problems and the compiler now works. Ask Anton Erasmus, He posts here it was his compiler. The problem was sorted out in Jan/Feb 2009

I have done similar for other dongled compilers for companies that have gone. IT is NOT the problem people make it out to be.

In your case...

A court might determine that it is all "software" This has not been tested in court for software but in other areas it has.

Maybe. As I said this is a real can of worms that will only make the lawyers rich.

Yes and the FSF takes them to court as has happened in some cases.

Yes. However very few companies cheat on commercial tools because most are dongled and it is too obviously illegal.

Fine but a lot, possibly the majority of commercial users don't release the source.

--
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\
\/\/\/\/\ Chris Hills  Staffs  England     /\/\/\/\/
\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/\/

Consider a branch in which unlawful copies dominate paid-for copies - PC games. I certainly have a few games which I have bought and paid for, and then downloaded cracks or cracked versions which I then play - because I have no interest in artificial restrictions and inconveniences such as having a particular disk in the drive while playing, or "phoning home" to supplier's website. How exactly do such restrictions benefit anyone?

As Dombo says, cheaters and free-loaders will always get a free copy (or else they won't use the software at all).

You ask others for evidence - do you have evidence for this "vast number" ? There are certainly examples of companies who have broken the GPL - many of whom did not really understand the problem, and corrected the issue when asked (sometimes forcefully). Some have been sued, and thereby forced to follow the legal requirements of the license (just like abusers of commercial licenses get sued). But I'm not sure I would even claim that a "vast number of people" modify the Open Source software they use - the majority simply use it.

And even among those that *do* modify it, most do so for their own use - there is no requirement to release the changes unless you distribute the software to others. I know I certainly have modified open source software for use by myself and specific customers - the changes are of no interest to anyone else, so there is no benefit in releasing them, and the customer gets the modified source along with the binary.

I think this is very unusual - but I think there is little doubt that ByteCraft *are* unusual in the effort they go to for customer service.

I only really know the details of one commercial compiler well enough for comparison, but in that case the compiler source could not be licensed because it contains third-party commercial code. As Chris said, I that applies to a lot of commercial software (for compilers or any other software).

Reliability problems in dedicated systems are almost always the fault of the application software, not the underlying system. Without more information here, it could be as simple as an easier user interface leading to less confusion.

You are right that sweeping statements are seldom completely true. As a sweeping statement, with a few exceptions, unix systems are far more reliable than windows systems, as are Linux, BSD and MacOS systems. The reputation for problems with Windows is often exaggerated, but it is not unfounded.

In the case mentioned here, the commercial tool had bugs. What exactly is the benefit of this commercial vendors "effective test suite"? Perhaps they can give you a nicely formatted report using an expensive independent test suite which shows exactly where their compiler fails?

I agree that test suites are important, obviously, but I think you overrate the benefits of your favourite test suites, and you overrate the amount of testing commercial developers do (I'm sure there are some that do it well - but very far from all), and you underestimate the amount of testing open source developers do (just like commercial developers, it will vary).

I follow a number of mailing lists, and often help out. It's on my own time, and I'm not paid for it. Sometimes I feel I've helped share a little wisdom around the world. Sometimes I feel that I've helped the tool's developers improve the software, which is then of benefit to me (and my employer). There are some people who don't measure everything, including all their time, in terms of money. And there are some people who are willing to contribute some of their time and money for the benefit of others without asking anything in return.

Again, I think you use terms like "most compiler vendors" rather loosely. And you omit the issue that many such vendors will charge significantly for the service (understandably, of course - but it's still worth mentioning).

Chris is a man on a mission. He is deathly afraid that somewhere someone is getting something from someone other than his preferred suppliers. And if they're getting it for free, well, that is the very nadir of depravity in his moral sphere.

Walter, you obviously have a track record of innovation in this industry that gives you first dibs on speaking on this topic, but I just don't see your point. Are you claiming that the progress in embedded technology in the Open Source era, which I would arbitrarily define as the last 10 years, was significantly impaired? I just don't see it this way.

The proliferation of standard platforms, whether it's TCP/IP or web or ARM or GCC etc. etc. enabled an amazing acceleration of novel, innovative things. Open Source can't claim the entire responsibility, but it certainly contributed to the overall climate of interoperability.

Contrarily, the areas with most proprietary barriers to entry, maintained by entrenched players, seem to be stuck on expensive and unimaginative. For example, cellular data: the big providers' idea of progress is SMS at

20 cents per 160 bytes (1250$/MB), whereas Google Android offers interesting geolocation apps and things like Shopping Assistant.

I remember this story from pre-Open Source era (early 80s). When a RISC vendor was trying to break into a new market, his reluctant customer asked: "So, your system is twice as fast as my existing hardware, and costs only half the money. Give me one good reason why I should switch to you".

The insight here is that to compete, there has to be a significant, compelling, knock-out advantage. The non-Open Source applications can gain market share, but they have to be really better and more innovative, not only initially but also in the long term.

--
		Przemek Klosowski, Ph.D.

... snip ...

First point - open source compilers may normally be districuted under GPC, but not GPL. GPL is specifically designed to allow use of libraries in software whose code is retained. However, to ensure the library can be revised and debugged the _using_ software must be distributed in a relinkable manner. This means such things as source code, but can be satisfied by linkable object code (and no source). Note that these restrictions are primarily designed to ensure the continued freedom of the open source product, not to restrict the users unnecessarily.

In general commercially distributed library source carries restrictions as to propagation. Thus it can never legally reach the actual user, who bought the software. These restrictions are not present in GPLd libraries.

--
 [mail]: Chuck F (cbfalconer at maineline dot net) 
 [page]: 
            Try the download section.

... snip ...

Which is one more advantage to using GNU licensed software. This way the 'small company' can complain to GNU, who probably has a better chance of resolving court cases.

--
 [mail]: Chuck F (cbfalconer at maineline dot net) 
 [page]: 
            Try the download section.