- About Scala
- Documentation
- Code Examples
- Software
- Scala Developers
Scala Meeting Report 2012/02/07
Tue, 2012-02-07, 18:07
This summary is intended for Scala contributors and maintainers, and assumes a certain familiarity with Scala's internals.
It is for your information only, provided “AS IS” with no warranties and it confers no rights (nor lefts).
Scala Meeting Report 2012/02/07
Attending
Adriaan: VirtPatMat (handling the catches in a try/catch)Alex: Parallel collections - CTries in trunk
Heather: doc-site tasks, dataflow with Alex, akka tasks, serializing closures
Martin: worked on inline classes (now called value classes), traits that inherit from Any, value class translation
Manohar: kmeans - big performance slowdown when using a single code block for the entire program - code size is also important -- play with the JVM parameters
Eugene: fixed reification bugs, got rid of the LiftCode phase, its logic has been transplanted into scala.reflect.macro.Context#reify
Christopher: slick bugfixes / refactorings, now using sbt for slick
Miguel: parallel inliner, exploring genJVM with ASM
Ingo: multi-way dependencies into scala react - ported a linear constraint solver to scala
Vlad: breaking nightlies, scaladoc patches, lms/delite work
Tiark: thesis writing
Hubert: fixed bug that was crashing sbt; type debugger
Lukas: finished design of lambda-calculus with “like”-effects, putting it in a tech report. After that, will start updating the compiler plugin and put things to work.
Georges: akka cluster runtime. Initial direction: io and data partitioning.
Discussion
Role of scaladoc in builds
1) let’s automate testing the full scaladoc suite for every commit2) let’s not let scaladoc fail the nightly (but that should become easier to avoid given (1))
Scala-lang reorganization
We're going to prune old content, simplify the structure. First step: getting statistics from drupal (list of all content, when it was last updated/accessed).Next step: crowdsource decision of which content to keep, and which to archiveThe doc site Following the wikipedia model, we'll have a banner for outdated/inaccurate content so people are aware of it.(When this becomes necessary. Next steps could include automating this process.)
Wed, 2012-02-08, 16:41
#2
Re: Scala Meeting Report 2012/02/07
среда, 8 февраля 2012 г. 0:06:30 UTC+7 пользователь Adriaan Moors написал:
Lukas: finished design of lambda-calculus with “like”-effects, putting it in a tech report.
Is there any chance for someone from the "outer world" to read it?
On this page there's only link to 11 pages of proofs :(
http://lampwww.epfl.ch/publications/
Regards,
Pavel.
Wed, 2012-02-08, 16:51
#3
Re: Re: Scala Meeting Report 2012/02/07
On 02/08/2012 04:39 PM, Pavel Pavlov wrote:
See header of Adriaan's email.
среда, 8 февраля 2012 г. 0:06:30 UTC+7 пользователь Adriaan Moors написал:Lukas: finished design of lambda-calculus with “like”-effects, putting it in a tech report.
Is there any chance for someone from the "outer world" to read it?
On this page there's only link to 11 pages of proofs :(
http://lampwww.epfl.ch/publications/
Regards,
Pavel.
See header of Adriaan's email.
Wed, 2012-02-08, 17:01
#4
Re: Re: Scala Meeting Report 2012/02/07
On 02/08/2012 04:50 PM, Hubert Plociniczak wrote:
I should have added "no pun intended" to avoid any angry emails. A slightly longer answer: we do not have necessary resources to elaborate on each description.
Of course Lukas is free to elaborate on that particular subject ;)
Thanks,
hubert
4F3299DB [dot] 20501 [at] epfl [dot] ch" type="cite"> On 02/08/2012 04:39 PM, Pavel Pavlov wrote:
среда, 8 февраля 2012 г. 0:06:30 UTC+7 пользователь Adriaan Moors написал:Lukas:
finished design of lambda-calculus with “like”-effects,
putting it in a tech report.
Is there any chance for someone from the "outer world" to read it?
On this page there's only link to 11 pages of proofs :(
http://lampwww.epfl.ch/publications/
Regards,
Pavel.
See header of Adriaan's email.
I should have added "no pun intended" to avoid any angry emails. A slightly longer answer: we do not have necessary resources to elaborate on each description.
Of course Lukas is free to elaborate on that particular subject ;)
Thanks,
hubert
Wed, 2012-02-08, 17:11
#5
Re: Re: Scala Meeting Report 2012/02/07
On Wednesday, 8. February 2012 at 16:39, Pavel Pavlov wrote:
no, not currently, i'm afraid..lukas
среда, 8 февраля 2012 г. 0:06:30 UTC+7 пользователь Adriaan Moors написал:Lukas: finished design of lambda-calculus with “like”-effects, putting it in a tech report.
Is there any chance for someone from the "outer world" to read it?
On this page there's only link to 11 pages of proofs :(
http://lampwww.epfl.ch/publications/
Wed, 2012-02-08, 17:21
#6
Re: Re: Scala Meeting Report 2012/02/07
On Wed, Feb 8, 2012 at 3:56 PM, Hubert Plociniczak <hubert.plociniczak@epfl.ch> wrote:
That much is understood. But there is no way to known which questions can or cannot be answered given existing resources. So, it's fair play to ask questions with the understanding that they may not be answered. Like any other question in the mailing list. I don't think we need reminders about that (unless someone complains about lack of answers).
Best,Ismael
I should have added "no pun intended" to avoid any angry emails. A slightly longer answer: we do not have necessary resources to elaborate on each description.
That much is understood. But there is no way to known which questions can or cannot be answered given existing resources. So, it's fair play to ask questions with the understanding that they may not be answered. Like any other question in the mailing list. I don't think we need reminders about that (unless someone complains about lack of answers).
Best,Ismael
Wed, 2012-02-08, 17:31
#7
Re: Re: Scala Meeting Report 2012/02/07
On Wed, Feb 8, 2012 at 5:03 PM, Ismael Juma <ismael@juma.me.uk> wrote:
On Wed, Feb 8, 2012 at 3:56 PM, Hubert Plociniczak <hubert.plociniczak@epfl.ch> wrote:
I should have added "no pun intended" to avoid any angry emails. A slightly longer answer: we do not have necessary resources to elaborate on each description.
That much is understood. But there is no way to known which questions can or cannot be answered given existing resources.
So, it's fair play to ask questions with the understanding that they may not be answered. Like any other question in the mailing list. I don't think we need reminders about that (unless someone complains about lack of answers).questions and feedback are welcome -- absolutely! apologies on behalf of the OP.
adriaan
Hey,
Ingo: multi-way dependencies into scala react - ported a linear
constraint solver to scala
Is this available somewhere?
On Tue, Feb 7, 2012 at 15:06, Adriaan Moors wrote:
> This summary is intended for Scala contributors and maintainers, and assumes
> a certain familiarity with Scala's internals.
>
> It is for your information only, provided “AS IS” with no warranties and it
> confers no rights (nor lefts).
>
> Scala Meeting Report 2012/02/07
>
> Attending
>
> Adriaan: VirtPatMat (handling the catches in a try/catch)
> Alex: Parallel collections - CTries in trunk
> Heather: doc-site tasks, dataflow with Alex, akka tasks, serializing
> closures
> Martin: worked on inline classes (now called value classes), traits that
> inherit from Any, value class translation
> Manohar: kmeans - big performance slowdown when using a single code block
> for the entire program - code size is also important -- play with the JVM
> parameters
> Eugene: fixed reification bugs, got rid of the LiftCode phase, its logic has
> been transplanted into scala.reflect.macro.Context#reify
> Christopher: slick bugfixes / refactorings, now using sbt for slick
> Miguel: parallel inliner, exploring genJVM with ASM
> Ingo: multi-way dependencies into scala react - ported a linear constraint
> solver to scala
> Vlad: breaking nightlies, scaladoc patches, lms/delite work
> Tiark: thesis writing
> Hubert: fixed bug that was crashing sbt; type debugger
> Lukas: finished design of lambda-calculus with “like”-effects, putting it in
> a tech report. After that, will start updating the compiler plugin and put
> things to work.
> Georges: akka cluster runtime. Initial direction: io and data partitioning.
>
> Discussion
>
> Role of scaladoc in builds
>
> 1) let’s automate testing the full scaladoc suite for every commit
> 2) let’s not let scaladoc fail the nightly (but that should become easier to
> avoid given (1))
>
> Scala-lang reorganization
>
> We're going to prune old content, simplify the structure.
> First step: getting statistics from drupal (list of all content, when it was
> last updated/accessed).
> Next step: crowdsource decision of which content to keep, and which to
> archive
>
> The doc site
> Following the wikipedia model, we'll have a banner for outdated/inaccurate
> content so people are aware of it.
> (When this becomes necessary. Next steps could include automating this
> process.)