www.digitalmars.com         C & C++   DMDScript  

digitalmars.D.announce - Brief report about the Tango Conference 2008 in Torun

reply Alexander Panek <alexander.panek brainsware.org> writes:
http://panic.brainsware.at/~jim/blog/2008/09/30/tango-conference-2008/

Digg: http://digg.com/programming/Tango_Conference_2008
Reddit: 
http://www.reddit.com/r/reddit.com/comments/74hn7/panic_tango_conference_2008/
Sep 30 2008
parent reply bearophile <bearophileHUGS lycos.com> writes:
From:
 http://panic.brainsware.at/~jim/blog/2008/09/30/tango-conference-2008/

Dr. Rafa&#322; Bocian showed us some of his students works written in the D &
Tango course,<

What does he/she thinks about using D to teach? What kind of level/kind of courses they are?
regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this? Bye and thank you, bearophile
Sep 30 2008
parent reply Alexander Panek <alexander.panek brainsware.org> writes:
bearophile wrote:
 From:
 http://panic.brainsware.at/~jim/blog/2008/09/30/tango-conference-2008/

 Dr. Rafa&#322; Bocian showed us some of his students’ works written in the D
& Tango course,<

What does he/she thinks about using D to teach? What kind of level/kind of courses they are?

He seems to like D and its flexibility. He didn’t explicitely talk about that, though. I’m not sure how to lable the level :) I think it was some sort of course for beginners in D but with at least some background in programming. They covered IO, networking, GUI et al, using Tango.
 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime. We didn’t really come to any conclusion because it still depends on Walter in some way and he wasn’t there. But apart from that, pretty much everyone was seeing the common runtime as the way to go. Does that answer your questions? :)
Sep 30 2008
next sibling parent reply bearophile <bearophileHUGS lycos.com> writes:
Alexander Panek:
 He seems to like D and its flexibility.

I hope to read a report or an article about the usage of D in teaching, with comparisons against Java too :-) Even if it says bad things regarding D, it can be very useful still. Another useful thing may be a list of the bugs put by students in their D programs. This list can be used to remove some spots that are currently bug-prone of D.
 Well, there’s a project on dsource (druntime: 
 http://dsource.org/projects/druntime)

The project says:
A freestanding version of the Tango runtime usable by Phobos and Tango<

Now we just need 'A freestanding version of the Phobos runtime usable by Tango and Phobos' and we are even ;o)
 Does that answer your questions? :)

Thank you, bearophile
Sep 30 2008
parent reply Tomas Lindquist Olsen <tomas famolsen.dk> writes:
bearophile wrote:
 Alexander Panek:
 He seems to like D and its flexibility.

I hope to read a report or an article about the usage of D in teaching, with comparisons against Java too :-) Even if it says bad things regarding D, it can be very useful still. Another useful thing may be a list of the bugs put by students in their D programs. This list can be used to remove some spots that are currently bug-prone of D.

He didn't really get much into that, what he did talk a lot about was some of the parts of the D spec that students had a hard time explaining/understanding.
 Well, there’s a project on dsource (druntime: 
 http://dsource.org/projects/druntime)

The project says:
 A freestanding version of the Tango runtime usable by Phobos and Tango<

Now we just need 'A freestanding version of the Phobos runtime usable by Tango and Phobos' and we are even ;o)

Thing is, Tango's runtime is much more self contained, IIRC it's been proven superior several times as well (as in faster). The main thing I got out of this was that a common interface for the D runtime was a big point. So nothing wrong with changing the Phobos runtime to implement this interface, it'd be a lot more work though for little to no gain... As mentioned there was a compiler/runtime workshop sunday morning. I found this one quite interesting. A lot of good points about flaws in the D spec (regarding future new compilers and portability), general mistakes in the language design etc. were "presented". I think the idea was to compile a list of the general conclusions, not sure who actually did that (if any). Though the notes should still exist. I'm sure these would spawn a very long NG thread when posted ;) -Tomas
Oct 01 2008
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
Tomas Lindquist Olsen wrote:
 As mentioned there was a compiler/runtime workshop sunday morning. I 
 found this one quite interesting. A lot of good points about flaws in 
 the D spec (regarding future new compilers and portability), general 
 mistakes in the language design etc. were "presented".

It would be great if some of those were posted here so that Walter knows. Andrei
Oct 01 2008
parent reply Walter Bright <newshound1 digitalmars.com> writes:
Andrei Alexandrescu wrote:
 Tomas Lindquist Olsen wrote:
 As mentioned there was a compiler/runtime workshop sunday morning. I 
 found this one quite interesting. A lot of good points about flaws in 
 the D spec (regarding future new compilers and portability), general 
 mistakes in the language design etc. were "presented".

It would be great if some of those were posted here so that Walter knows.

I agree. I can't do anything about it otherwise.
Oct 01 2008
parent reply =?UTF-8?B?QWxleGFuZGVyIFDDoW5law==?= writes:
Walter Bright wrote:
 Andrei Alexandrescu wrote:
 Tomas Lindquist Olsen wrote:
 As mentioned there was a compiler/runtime workshop sunday morning. I 
 found this one quite interesting. A lot of good points about flaws in 
 the D spec (regarding future new compilers and portability), general 
 mistakes in the language design etc. were "presented".

It would be great if some of those were posted here so that Walter knows.

I agree. I can't do anything about it otherwise.

Some people are just assembling a list as we speak. There will be a seperate thread for that, as soon as it is finished.
Oct 01 2008
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
Alexander Pánek wrote:
 Walter Bright wrote:
 Andrei Alexandrescu wrote:
 Tomas Lindquist Olsen wrote:
 As mentioned there was a compiler/runtime workshop sunday morning. I 
 found this one quite interesting. A lot of good points about flaws 
 in the D spec (regarding future new compilers and portability), 
 general mistakes in the language design etc. were "presented".

It would be great if some of those were posted here so that Walter knows.

I agree. I can't do anything about it otherwise.

Some people are just assembling a list as we speak. There will be a seperate thread for that, as soon as it is finished.

Terrific! It would actually be cool if separate threads would be started for each of these issues so as to compartmentalize the discussions. Andrei
Oct 01 2008
parent Walter Bright <newshound1 digitalmars.com> writes:
Andrei Alexandrescu wrote:
 Alexander Pánek wrote:
 Walter Bright wrote:
 Andrei Alexandrescu wrote:
 Tomas Lindquist Olsen wrote:
 As mentioned there was a compiler/runtime workshop sunday morning. 
 I found this one quite interesting. A lot of good points about 
 flaws in the D spec (regarding future new compilers and 
 portability), general mistakes in the language design etc. were 
 "presented".

It would be great if some of those were posted here so that Walter knows.

I agree. I can't do anything about it otherwise.

Some people are just assembling a list as we speak. There will be a seperate thread for that, as soon as it is finished.

Terrific! It would actually be cool if separate threads would be started for each of these issues so as to compartmentalize the discussions.

Yes, one omnibus posting will likely be a mess.
Oct 01 2008
prev sibling parent reply Sean Kelly <sean invisibleduck.org> writes:
Alexander Panek wrote:
 bearophile wrote:
 From:

 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

I just thought I'd mention that if anyone wants to play with this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now. Sean
Oct 03 2008
next sibling parent reply dsimcha <dsimcha yahoo.com> writes:
== Quote from Sean Kelly (sean invisibleduck.org)'s article
 Alexander Panek wrote:
 bearophile wrote:
 From:

 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now. Sean

So what's the status, then, of Tango for D2? I assume that having Phobos and Tango be able to be used together in D2 is kind of the point of druntime.
Oct 03 2008
parent Sean Kelly <sean invisibleduck.org> writes:
dsimcha wrote:
 == Quote from Sean Kelly (sean invisibleduck.org)'s article
 Alexander Panek wrote:
 bearophile wrote:
 From:
 regarding a common runtime for Phobos and Tango,<


http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

So what's the status, then, of Tango for D2? I assume that having Phobos and Tango be able to be used together in D2 is kind of the point of druntime.

The user code will have to be updated, and that /may/ be enough of a change that a branch will be required. If you're interested in using Tango with D 2.0 for now I'd suggest using DSSS instead of trying to build the user library, and just see how it goes. My focus is on the runtime, so I honestly can't say how far you'll get. Sean
Oct 03 2008
prev sibling parent reply =?UTF-8?B?QWxleGFuZGVyIFDDoW5law==?= writes:
Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:

 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

I just thought I'd mention that if anyone wants to play with this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Ah, good to know. What are the differences between druntime and Tango’s runtime? What about the features Tango has but Phobos doesn’t?
Oct 03 2008
parent reply Sean Kelly <sean invisibleduck.org> writes:
Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:

 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

I just thought I'd mention that if anyone wants to play with this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Ah, good to know. What are the differences between druntime and Tango’s runtime? What about the features Tango has but Phobos doesn’t?

The location of things have changed and certain bits have been tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean
Oct 03 2008
parent reply Jason House <jason.james.house gmail.com> writes:
Sean Kelly Wrote:

 Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:

 regarding a common runtime for Phobos and Tango,<

Can you give a little summary of this?

Well, there’s a project on dsource (druntime: http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

I just thought I'd mention that if anyone wants to play with this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Ah, good to know. What are the differences between druntime and Tango’s runtime? What about the features Tango has but Phobos doesn’t?

The location of things have changed and certain bits have been tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean

What has been done to coordinate with Walter and/or the Phobos development team? Does the license match Walter's requirement? Are there any known reasons why D2 would not want to adopt this runtime?
Oct 03 2008
parent reply Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
Jason House wrote:
 Sean Kelly Wrote:
 
 Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:
 regarding a common runtime for Phobos and Tango,<


http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Tango’s runtime? What about the features Tango has but Phobos doesn’t?

tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean

What has been done to coordinate with Walter and/or the Phobos development team? Does the license match Walter's requirement? Are there any known reasons why D2 would not want to adopt this runtime?

D2 will in all likelihood adopt Sean's runtime. We only need to put in the adaptation work. I think Sean has done a tremendous contribution to the community. Andrei
Oct 03 2008
next sibling parent reply =?UTF-8?B?QWxleGFuZGVyIFDDoW5law==?= writes:
Andrei Alexandrescu wrote:
 Jason House wrote:
 Sean Kelly Wrote:

 Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:
 regarding a common runtime for Phobos and Tango,<


http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Tango’s runtime? What about the features Tango has but Phobos doesn’t?

tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean

What has been done to coordinate with Walter and/or the Phobos development team? Does the license match Walter's requirement? Are there any known reasons why D2 would not want to adopt this runtime?

D2 will in all likelihood adopt Sean's runtime. We only need to put in the adaptation work. I think Sean has done a tremendous contribution to the community.

Any thoughts on using druntime in D1?
Oct 03 2008
parent Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
Alexander Pánek wrote:
 Andrei Alexandrescu wrote:
 Jason House wrote:
 Sean Kelly Wrote:

 Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:
 regarding a common runtime for Phobos and Tango,<


http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Tango’s runtime? What about the features Tango has but Phobos doesn’t?

tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean

What has been done to coordinate with Walter and/or the Phobos development team? Does the license match Walter's requirement? Are there any known reasons why D2 would not want to adopt this runtime?

D2 will in all likelihood adopt Sean's runtime. We only need to put in the adaptation work. I think Sean has done a tremendous contribution to the community.

Any thoughts on using druntime in D1?

Walter needs to answer about D1. Andrei
Oct 03 2008
prev sibling next sibling parent reply dsimcha <dsimcha yahoo.com> writes:
== Quote from Andrei Alexandrescu (SeeWebsiteForEmail erdani.org)'s
 D2 will in all likelihood adopt Sean's runtime. We only need to put in
 the adaptation work. I think Sean has done a tremendous contribution to
 the community.
 Andrei

Is the new model, then, going to become that Phobos and Tango are strictly user-level libraries that both run on top of druntime and can therefore be arbitrarily mixed and matched in code? I think this would do wonders, not only for de-fragmenting the D community but for the development of both libraries. It seems that Phobos and Tango had been converging to a degree in terms of feature set, etc., with lots of duplication of effort. This is likely because using them in the same code base had been mutually exclusive, so everyone using one lib wanted that one killer feature from the other. Now, with mixing and matching supported, Phobos can just be what it was designed to be, namely a compact standard library with only stuff that's very general purpose and a more procedural and/or functional style, and Tango can be what it was designed to be, a more Java-style standard library with just about everything and a more OO style.
Oct 03 2008
parent Andrei Alexandrescu <SeeWebsiteForEmail erdani.org> writes:
dsimcha wrote:
 == Quote from Andrei Alexandrescu (SeeWebsiteForEmail erdani.org)'s
 D2 will in all likelihood adopt Sean's runtime. We only need to put in
 the adaptation work. I think Sean has done a tremendous contribution to
 the community.
 Andrei

Is the new model, then, going to become that Phobos and Tango are strictly user-level libraries that both run on top of druntime and can therefore be arbitrarily mixed and matched in code?

That is the case about Phobos2 at least. Andrei
Oct 03 2008
prev sibling parent Jason House <jason.james.house gmail.com> writes:
Andrei Alexandrescu Wrote:

 Jason House wrote:
 Sean Kelly Wrote:
 
 Alexander Pánek wrote:
 Sean Kelly wrote:
 Alexander Panek wrote:
 bearophile wrote:
 From:
 regarding a common runtime for Phobos and Tango,<


http://dsource.org/projects/druntime) which is supposed to hold that common runtime.

this, the runtime is fully functional and there are versions available for both D 1.0 and D 2.0 (in a branch and trunk, respectively). Assuming everything works out however, I expect that only Phobos for D 2.0 will actually use the common runtime, since D 1.0 is in maintenance mode now.

Tango’s runtime? What about the features Tango has but Phobos doesn’t?

tightened up a bit compared to Tango, but the functionality should be identical to Tango's runtime. With that in mind, if you want to try it out, build druntime and replace tango-base-dmd.lib with druntime-dmd.lib, get rid of the .di files in tango/core that are generated by the lib build process, and use the top-level memory, thread, etc, modules in their place. The project is really just a proof of concept at this point. I've got a version of Phobos for D 2.0 working on top of druntime with no problems, for example, but there is no official support yet for druntime from Phobos or Tango. In short, druntime works just fine, but to actually use it right now you're pretty much on your own. I'll be happy to answer any questions regarding setup, but don't have any plans yet for installation scripts, packaging, etc. By the way, DMD is the only supported compiler at the moment. I'll add support for the others later if the project actually takes off. Sean

What has been done to coordinate with Walter and/or the Phobos development team? Does the license match Walter's requirement? Are there any known reasons why D2 would not want to adopt this runtime?

D2 will in all likelihood adopt Sean's runtime. We only need to put in the adaptation work. I think Sean has done a tremendous contribution to the community. Andrei

I'm really pushing my luck now, but is there any timeline for finalizing D2's evaluation/adoption of druntime? My understanding from Sean's post is that he's already adapted Phobos to use the shared runtime. It sounds like it'd be really easy to drop in.
Oct 03 2008