FeepingCreature Wrote:

> Phobos1 is shit. The Tango devs know this, the Phobos devs know it. Anyone who denies it has never compared the Phobos and Tango sourcecode.

It's impossible to verify those claims because reading the Tango source might taint one's mind and after that one wouldn't be allowed to contribute any code to Phobos anymore.

> Your "simple" solution is never gonna happen. You're not freaking Alexander the Great, cutting the Gordian Knot. The way D2 is going is the best solution for both sides, imho; but _anything_ that prevents Tango/Phobos interop in D2, or pushes away Tango devs, or pushes away Phobos devs - should be treated as a *severe* threat to the future of the language. We *absolutely need* to present a unified front in D2. We fucked this up once already; let's not repeat that experience.

The Tango developers could have handed over all copyrights to Walter or Phobos. This would solve the licensing problems if anything needs to change later. Many open source projects such as MySQL do this.

Instead they yearned the attribution. Which one is more important, personal fame or potential solid enterprise support? If the library isn't rocket science or doesn't cure the cancer, what value does the attribution have then? The new Phobos licensing is altruistic, it reflects the modest mentality of the contributors.

On 30.04.2010 16:04, lurker wrote:
> The Tango developers could have handed over all copyrights to Walter or Phobos. This would solve the licensing problems if anything needs to change later. Many open source projects such as MySQL do this.
> 

They could have jumped off a bridge too. Yay, no more Tango. All problems magically go away.

Well except the Tango devs' problems, but who cares about those.

> Instead they yearned the attribution.

Oh, because you always make perfect license choices on first try in a muddy context, or alternately when it turns out you made the wrong choice, you can always change the license without hassle! Because it's not like there's other people who contributed code that you can't reach, that never happens.

Sometimes I forget you're a superhuman fantasy creature.

lurker wrote:
> FeepingCreature Wrote:
> 
>> Phobos1 is shit. The Tango devs know this, the Phobos devs know it. Anyone who denies it has never compared the Phobos and Tango sourcecode.
> 
> It's impossible to verify those claims because reading the Tango source might taint one's mind and after that one wouldn't be allowed to contribute any code to Phobos anymore.

Well, there are other objective means.

This is a subjective statement: but as someone who has used D extensively over the past several years, including both Phobos and Tango, I honestly believe that Tango is generally of a higher quality.

Except for Tango's Zip code which is an abomination and should be killed with fire--the original author is clearly a talentless hack.

>> Your "simple" solution is never gonna happen. You're not freaking Alexander the Great, cutting the Gordian Knot. The way D2 is going is the best solution for both sides, imho; but _anything_ that prevents Tango/Phobos interop in D2, or pushes away Tango devs, or pushes away Phobos devs - should be treated as a *severe* threat to the future of the language. We *absolutely need* to present a unified front in D2. We fucked this up once already; let's not repeat that experience.
> 
> The Tango developers could have handed over all copyrights to Walter or Phobos. This would solve the licensing problems if anything needs to change later.

I don't know how many times this has to be explained.

To quote myself:

"Thirdly, the Tango maintainers have *ALREADY TRIED* to change Tango's license.  They wanted to move to just Apache 2.0 on the basis that it was similar enough to the AFL to allow this without too much trouble.

"The problem was that of the 50-odd contributors, there are people who they simply couldn't get in contact with.  Without express permission, they *CANNOT* legally change the license to something incompatible."

> Many open source projects such as MySQL do this.

(Aside: I find it somewhat amusing that you're suggesting the Tango devs should relinquish all claim on their work; the same thing the FSF asked for in order to include the GDB patches.)

> Instead they yearned the attribution. Which one is more important, personal fame or potential solid enterprise support? If the library isn't rocket science or doesn't cure the cancer, what value does the attribution have then?

We've already established that this is a legal issue, not one of ego. It'd be nice if you refrained from personal attacks.

> The new Phobos licensing is altruistic, it reflects the modest
mentality of the contributors.

The Boost license still requires source to contain attribution.

Lars commented in his post that he doesn't like the binary attribution requirement.  But he's stuck with it because of the code's heritage.

As I tried very hard to explain, this is not about attempting to sabotage D or Phobos or, for that matter, anyone or anything.

Please, PLEASE stop with the needless rhetoric and hate.

On 30.04.2010 16:27, Daniel Keep wrote:
> 
> 
> lurker wrote:
>> FeepingCreature Wrote:
>>
>>> Phobos1 is shit. The Tango devs know this, the Phobos devs know it. Anyone who denies it has never compared the Phobos and Tango sourcecode.
>>
>> It's impossible to verify those claims because reading the Tango source might taint one's mind and after that one wouldn't be allowed to contribute any code to Phobos anymore.
> 
> Well, there are other objective means.
> 
> This is a subjective statement: but as someone who has used D extensively over the past several years, including both Phobos and Tango, I honestly believe that Tango is generally of a higher quality.
> 
> Except for Tango's Zip code which is an abomination and should be killed with fire--the original author is clearly a talentless hack.
> 
I don't know the Tango zip code, but -- the Phobos zip code has/had a very common bug where it causes/caused silent data loss. Beat that.

Daniel Keep wrote:

...
> Except for Tango's Zip code which is an abomination and should be killed with fire--the original author is clearly a talentless hack.

Well, at least he has some character!

...
> 
> As I tried very hard to explain, this is not about attempting to sabotage D or Phobos or, for that matter, anyone or anything.
> 
> Please, PLEASE stop with the needless rhetoric and hate.

Yes please, thanks for your posts. The situation is rather unfortunate, tragic even. It doesn't need to get worse than that!

Daniel Keep wrote:
> 
> lurker wrote:
>> The Tango developers could have handed over all copyrights to Walter or Phobos. This would solve the licensing problems if anything needs to change later.
> 
> I don't know how many times this has to be explained.
> 
> To quote myself:
> 
> "Thirdly, the Tango maintainers have *ALREADY TRIED* to change Tango's
> license.  They wanted to move to just Apache 2.0 on the basis that it
> was similar enough to the AFL to allow this without too much trouble.
> 
> "The problem was that of the 50-odd contributors, there are people who
> they simply couldn't get in contact with.  Without express permission,
> they *CANNOT* legally change the license to something incompatible."

That's true, but largely irrelevant. Individual developers can make agreements about relicensing of their personal contributions, and stating that they're happy with their code being used in Phobos. Sean, Steven, and I did. AFAIK the other Tango developers have not. Everything's in version control, you can see who's contributed to which components. Sure, there'll be places where a dozen uncontactable people have been involved. But that shouldn't be an argument for making no progress.

It seems very clear to me that there are Tango developers who do not want any of their code to be used in Phobos. Which is fine, that's their choice. But I wish they'd have the decency to say so, so that the community stops wasting time on the issue.

I've tried for the past two years to make tiny steps towards unity. But Tango does not seem to be interested.

Please tell me I'm wrong.

On 30.04.2010 17:10, Don wrote:
> It seems very clear to me that there are Tango developers who do not want any of their code to be used in Phobos. Which is fine, that's their choice. But I wish they'd have the decency to say so, so that the community stops wasting time on the issue.
> 

So what you're saying is, you have this knowledge despite the relevant Tango devs not actually saying anything in that direction.

Could you maybe explain how you came to that conclusion, please?

== Quote from Don (nospam@nospam.com)'s article
> That's true, but largely irrelevant. Individual developers can make agreements about relicensing of their personal contributions, and stating that they're happy with their code being used in Phobos.

Walter said, basically, that since it's possible that SHOO may have used code from Tango, Tango devs should relicense their work. That's insulting. It's admitting theft and demanding that the victim call it a gift. If it were a policy, Walter would have a way of badgering us into relicensing most of Tango against our will.

I'm not saying that SHOO copied any Tango code. Walter's reaction, though, means I would never relicense any code for Phobos.

Chris Wright wrote:
> Walter said, basically, that since it's possible that SHOO may have used
> code from Tango, Tango devs should relicense their work.

Not exactly. To rephrase, I said that since SHOO has viewed Tango's source code, there is the appearance of impropriety. Not that there actually is any impropriety. It's the appearance I wish to avoid. I am not accusing anyone of infringement, and have no basis to.

I have asked the Tango devs to relicense their work. I feel that if that can be accomplished, this would bury this issue once and for all, and the rift between the communities should heal.

> That's
> insulting. It's admitting theft and demanding that the victim call it a
> gift. If it were a policy, Walter would have a way of badgering us into
> relicensing most of Tango against our will.

I can't make you do anything you don't want to. I especially have no means, desire, or intention of forcing anyone to change their license or give up their copyrights.


> I'm not saying that SHOO copied any Tango code. Walter's reaction,
> though, means I would never relicense any code for Phobos.

I've repeatedly stated, and say so again, that I give explicit permission to Tango to incorporate any or all of code I have written for Phobos into Tango, and to relicense those derived works as necessary to be compatible with Tango. Tango's garbage collector is such a derived work, and I have no issue with it.

As for Phobos code I did not write, in order to relicense it, you'd have to get the permission of the author(s) of it, which is stated in each module. But it is entirely unnecessary to relicense it - the Boost license allows you to use it any way you want to. The Boost license is not viral, it will not "infect" anything you hook it up to (neither does the BSD license - in fact, the only real difference between the BSD and Boost licenses is the binary attribution clause).

== Quote from Chris Wright (dhasenan@gmail.com)'s article
> == Quote from Don (nospam@nospam.com)'s article
> > That's true, but largely irrelevant. Individual developers can make agreements about relicensing of their personal contributions, and stating that they're happy with their code being used in Phobos.
> Walter said, basically, that since it's possible that SHOO may have used
> code from Tango, Tango devs should relicense their work. That's
> insulting. It's admitting theft and demanding that the victim call it a
> gift. If it were a policy, Walter would have a way of badgering us into
> relicensing most of Tango against our will.
> I'm not saying that SHOO copied any Tango code. Walter's reaction,
> though, means I would never relicense any code for Phobos.

http://www.google.com/search?sourceid=chrome&ie=UTF-8&q=chris+wright+site:tango.dsource.org

The loss is unbearable.


--
Gurney Halleck