My resignation as chair of SG5

[Michael Wong]

Hi all, at today's SAN meeting, I asked to resign as Chair of SG5, because of my increasing travel and my new assignment as chair of SG19 Machine Learning.

Let me say its been an honour to work with everyone here and I have gained an education from the best in the world on TM, unparalleled in any graduate School. Its been an absolute pleasure to represent this group.

In making sure that the group continues, I have negotiated with Hans Boehm to take over as chair, and it will be a pleasure to work under his direction as I have always been a fan of his handling of SG1.

Finally, the webex meeting booking, as I usually do, only lasted from DST to DST and terminated 2 weeks ago. It will be up to Hans to book the method as he likes.

It is sad to write this final report after chairing this great group (of friends), for the last ... 6-8 years? I can't even remember. I will continue to lurk. Since leaving IBM 3 years ago,  I have had less to do with TM, but continued because I wanted the structure to continue as I feel it is important. I will continue to lurk as I try to make sense of this thing called Machine Learning as required by my current job for the last three years.

[Justin Gottschlich]

Hi Michael -

I've been lurking, too. :) 

I'm sorry to see you leave the chair position, but I know that Hans will do an amazing job. It was great working with you and watching you lead SG5 from afar. I still remember the day we pitched SG5 and then unofficially started forming the group. I think that was back in 2011-2012? Crazy!

Speaking of machine learning (my primary focus of research these days), we should sync. I'm curious to know more about SG19 and might be interested in helping with that effort. I'll ping you in another thread. If you're going to be at NIPS, please come see our spotlight talk on Tuesday about anomaly detection and let's spend some time catching up! :)

Best,

Justin

--

Justin Gottschlich, Ph.D. ( http://justingottschlich.com )

Senior Staff Research Scientist, Intel Labs

Artificial Intelligence Lead, Programming Systems Research

Principal Investigator and Co-Founder, Intel/NSF CAPA Research Center

--
You received this message because you are subscribed to the Google Groups "SG5 - Transactional Memory" group.
To unsubscribe from this group and stop receiving emails from it, send an email to tm+unsu...@isocpp.org.
To post to this group, send email to t...@isocpp.org.
Visit this group at https://groups.google.com/a/isocpp.org/group/tm/.

[Michael Wong]

Thanks Justin missed you too. This group honestly self leads.
I plan to be at NIPS with perhaps a cross pond talk at Munich ESE Congress that week

While talking about past chairs, I just met Tatiana Speiseman giving a talk at LLVM and she and C lattner are at Deepmind. We hope to all work together again.

[Victor Luchangco]

Thanks, Michael, for your service to and leadership of SG5, and best wishes in your new roles.

- Victor

[Scott, Michael]

Hear, hear.  We’ll miss your leadership, but I’m glad to know you’ll be lurking!

- Michael

[Hans Boehm]

Thank you, Michael!

Unless someone made substantial progress they want to talk about, I suggest we cancel today's meeting. That will also give me time to set up an alternate teleconference arrangement and recover from the WG21 meeting.

By default, I will try to set up something based on Google Hangouts, though I don't yet have experience doing that with similar constraints. If you have a preference please let me know.

Since we've had issues getting critical mass for the last few meetings, I'd like to reduce the meeting schedule to once every four weeks by default, and see how that goes. If there is something urgent to discuss, we can revisit. So let's aim for a meeting on the 26th (Monday after Thanksgiving) in this time slot, and then probably every four weeks after that.

Possibly relevant news from the WG21 meeting for those who weren't there (others may want to add):

- The core part of the executors proposal is still/again possibly on track for C++20.

- Memory model fixes loosely based on the Lahav et al paper (Weaken seq_cst/acq/rel interaction, fix seq consistent fences, remove same-thread writes from release sequence) made it in. (P0668 & PP0982)

- People liked the asymmetric fences proposal (P1202), though many of us had hoped this could be done with less violence to the memory model.

Hans

[Scott, Michael]

On Nov 12, 2018, at 2:16 PM, Hans Boehm <bo...@acm.org> wrote:

Thank you, Michael!

Unless someone made substantial progress they want to talk about, I suggest we cancel today's meeting. That will also give me time to set up an alternate teleconference arrangement and recover from the WG21 meeting.

By default, I will try to set up something based on Google Hangouts, though I don't yet have experience doing that with similar constraints. If you have a preference please let me know.

Since we've had issues getting critical mass for the last few meetings, I'd like to reduce the meeting schedule to once every four weeks by default, and see how that goes. If there is something urgent to discuss, we can revisit. So let's aim for a meeting on the 26th (Monday after Thanksgiving) in this time slot, and then probably every four weeks after that.

Note that this puts our second meeting on Christmas Eve, which is probably suboptimal :-)

- Michael

[Hans Boehm]

The next meeting will be on Monday, 12:00 PST (same as before, just specified in a different time zone).

Unlike before, the new meeting coordinates will be:

meet.google.com/sbj-cvgh-vnd

or 

+1 208-925-0196‬ PIN: ‪255 542‬#

If there are any issues, please send me mail.

As Michael Scott points out, clearly one of the discussion points will need to be timing of the next meeting.

Hans

[Michael Wong]

Hi all, thanks for the kind words. I have now made Hans Boehm the Manager of this mailing list.

I assume the meeting is today.Please let me know anything I can help with.

[Hans Boehm]

Yes, the meeting is in half an hour.

Bassed on the last minutes I could find, I believe the minutes rota is:

Maged, Jens, Michael Scott, Michael Spear, Michael W, Hans, Victor

I expect we will generally talk about:

1. How to move TM lite forward.

2. Meeting schedule.

meet.google.com/sbj-cvgh-vnd

or 

+1 208-925-0196‬ PIN: ‪255 542‬#

[Tony V E]

Maybe it's time to change the email subject line, or start a new thread, I can't take more reminders of the sadness of Michael not being chair.

Although I appreciate Hans taking the wheel of course!

Be seeing you,

Tony