Texas LinuxFest 2011 call for papers now open

[Daniel Gibson]

Am 20.01.2011 21:15, schrieb Lutger Blijdestijn:
> Daniel Gibson wrote:
> ...
>
>> So yes, the point that D may cause less trouble than Java/.net can be
>> made, but you probably shouldn't claim that D doesn't infringe any
>> patents, because you can't possibly know (nobody can, there are just too
>> many software patents to check, even for big companies).
>>
>

 > Technically that is right, but I find it a bit of an understatement because
 > every non trival software project has potential issues. With .NET and Java
 > you *know* you have patent issues, with D any potential patent issue is a
 > tragic mistake that still has to be proven to exist. Those are not on the
 > same scale, so I wouldn't use the term 'less trouble' You also have
 > ownership to take into account, I would rather trust Walter Bright not using
 > submarine patent traps than MS or Oracle :)

I mostly agree with you :-)
Maybe "less trouble" is not the right word.. my point is: I guess Walter did not 
explicitly try to avoid patents when developing D (and, as mentioned before, you 
can't be sure anyway), so you probably can't advertise D as a patent-free 
language (like ogg vorbis is claimed to be a patent-free audio codec).
And I don't believe Walter is using any patent traps (Does he hold any patents? 
I wouldn't be surprised, and I don't really care because I don't believe he'd 
use it offensively) - but once D is popular you never know if a third party 
starts trolling. But of course, this may happen with any language or software in 
general. Furthermore, there was this thread in d.D about two Borland patents 
that are now owned by M$ that *may* be infringed by D...


> Perhaps I should elaborate a bit. mono is simply out of the question in a
> large part of linux. Fedora for example, has D as a feature for its 14
> release but doesn't support mono. Java is more complex, but if we take it
> out of the picture it leaves (some of) linux with C / C++ on the one hand
> and a lot of higher level dynamic languages on the other. In between are
> some more 'exotic' languages such as haskell. Perhaps I'm wrong, but I see a
> big void there where D can step in, mostly because of its set of features.
> In the non-open-source world, Java and .NET are already taking care of much
> of this void.

I agree.
A free language that works well with C and is at least as "user-friendly" as 
Java definitely has a lot of potential in the open source and esp. Linux/*BSD 
world. Many Open Source developers will prefer it to Java/Mono, because it 
doesn't use a VM (some may still complain about the garbage collector, though, 
but in the end they'd benefit from it, because it mostly prevents the quite 
common memory leaks).

Another thing that should be considered: For D2 to be really interesting in the 
Linux world, stream and socket handling needs to improve.
But it seems like this will happen, Andrei started a discussion about a new 
std.stream in d.D recently.

Cheers,
- Daniel