From dreyer at tti-c.org Fri Feb 2 15:24:42 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Fri Feb 2 15:24:45 2007 Subject: [Hyper] Reading for next Thursday Message-ID: <7fa251b70702021324s7a9ee61ek6f76f9f5b01943d3@mail.gmail.com> For next Thursday (meeting at 11:30 AM, probably in Ryerson 255), please read: Chapter 2 of my PhD thesis: http://tti-c.org/dreyer/thesis/chapter2.pdf The upside of the chapter is that it's totally informal, and is IMO a much better (or at least more comprehensive) explanation than the one we gave in our POPL paper. (The ways in which it differs from the POPL paper are discussed in detail at the end of the chapter.) The downside is that it's totally informal, so maybe I'll complement it by flashing some titillating typing rules at the meeting. Derek From jhr at cs.uchicago.edu Fri Feb 2 17:14:03 2007 From: jhr at cs.uchicago.edu (John Reppy) Date: Fri Feb 2 17:14:03 2007 Subject: [Hyper] next week's meeting Message-ID: I've switched our reservation for Ry255 to thursday for next week. - John From dreyer at tti-c.org Thu Feb 8 11:01:16 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Thu Feb 8 11:01:22 2007 Subject: [Hyper] Reminder: meeting today at 11:30 AM Message-ID: <7fa251b70702080901l177c86b5s35e92c9ec2c0aa35@mail.gmail.com> Perhaps this reminder comes a bit late, but we are meeting at Ryerson 255 at 11:30 AM today. Derek From dreyer at tti-c.org Fri Feb 9 16:16:59 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Fri Feb 9 16:17:04 2007 Subject: [Hyper] Reading for next week: More Dreyer! Message-ID: <7fa251b70702091416j2a4b705cg9b83cad7a1b9481b@mail.gmail.com> For next Tuesday, please read the journal version of my ICFP05 paper, "Recursive Type Generativity": http://ttic.uchicago.edu/~dreyer/papers/dps/jfp-revised.pdf This includes some technical improvements over the conference version, but don't feel obliged to read the whole thing. If you want to skip something, skip the technical "Meta-theory" section (Section 4), which is the main thing that I expanded from the conference paper. If you take that section out, it's probably the same length as the conference version, maybe even shorter. At the meeting, I will sketch for you how to scale this type system to the level of an internal-language module type system (it's pretty simple), and then the next week we will look at a hybrid Definition/Harper-Stone formalism for elaborating modules (recursive or not) into that type system. Derek From dreyer at tti-c.org Fri Feb 16 15:23:08 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Fri Feb 16 15:25:02 2007 Subject: [Hyper] Next week: Recursive modules, continued Message-ID: <7fa251b70702161323g7d87ab30k765ad8df75cf7dff@mail.gmail.com> Next Tuesday will be our final HYPER week looking at (my) work on type systems for modules. Please read my draft paper: "Practical Type Theory for Recursive Modules" http://ttic.uchicago.edu/~dreyer/papers/bimod/main.pdf For those who read my JFP paper last week, you can skip some significant chunks of this, although you may want to skim them... For those who don't want to read the paper too carefully, you might want to read the intro and the related work, and then search through the paper to figure out what those sections are talking about. (This should not require understanding all the elaboration rules, although I strongly urge you to take a look at those as well.) I'm sure there will be questions about the elaboration rules anyway, so I'll probably just go through them one by one on Tuesday. Derek From jhr at cs.uchicago.edu Mon Feb 26 14:16:29 2007 From: jhr at cs.uchicago.edu (John Reppy) Date: Mon Feb 26 14:31:13 2007 Subject: [Hyper] Nested data-parallel reading Message-ID: This week we will start looking at several papers on the topic of nested data parallelism. The first paper discusses this programming model from a programmer's point of view. Programming Parallel Algorithms Guy Blelloch CACM, 1996 It is available in HTML at http://www.cs.cmu.edu/~scandal/cacm.html and attached below. The second paper will be the DAMP paper on Data- parallel Haskell, which focuses more on implementation issues and is also attached below. - John -------------- next part -------------- A non-text attachment was scrubbed... Name: p85-blelloch.pdf Type: application/pdf Size: 836394 bytes Desc: not available Url : http://ttic.uchicago.edu/pipermail/hyper/attachments/20070226/a4eb3098/p85-blelloch-0001.pdf -------------- next part -------------- A non-text attachment was scrubbed... Name: chakravarty.pdf Type: application/pdf Size: 123014 bytes Desc: not available Url : http://ttic.uchicago.edu/pipermail/hyper/attachments/20070226/a4eb3098/chakravarty-0001.pdf From dreyer at tti-c.org Mon Feb 26 14:31:18 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Mon Feb 26 14:39:19 2007 Subject: [Hyper] Nested data-parallel reading In-Reply-To: References: Message-ID: <7fa251b70702261231t1861fef9lb8f88835dbf019f0@mail.gmail.com> The paper is available in PDF here: http://portal.acm.org/citation.cfm?id=227246&coll=portal&dl=ACM Derek On 2/26/07, John Reppy wrote: > This week we will start looking at several papers on the topic of > nested data parallelism. > The first paper discusses this programming model from a programmer's > point of view. > > Programming Parallel Algorithms > Guy Blelloch > CACM, 1996 > > It is available in HTML at > > http://www.cs.cmu.edu/~scandal/cacm.html > > and attached below. The second paper will be the DAMP paper on Data- > parallel > Haskell, which focuses more on implementation issues and is also > attached below. > > - John > > > > _______________________________________________ > Hyper mailing list > Hyper@ttic.uchicago.edu > http://ttic.uchicago.edu/mailman/listinfo/hyper > > > From dreyer at tti-c.org Mon Feb 26 14:50:37 2007 From: dreyer at tti-c.org (Derek Dreyer) Date: Mon Feb 26 14:58:34 2007 Subject: [Hyper] Nested data-parallel reading In-Reply-To: <7fa251b70702261231t1861fef9lb8f88835dbf019f0@mail.gmail.com> References: <7fa251b70702261231t1861fef9lb8f88835dbf019f0@mail.gmail.com> Message-ID: <7fa251b70702261250o4c986969n6878e59863375af4@mail.gmail.com> I assume that, as discussed last week, we will read this for Thursday (and move the HYPER meeting to Thursday permanently). If there are any objections to moving to Thursday, then I would suggest we skip meeting this week and read Blelloch's article for next Tuesday so that we have time to read it. Meeting tomorrow is too soon. Derek On 2/26/07, Derek Dreyer wrote: > The paper is available in PDF here: > > http://portal.acm.org/citation.cfm?id=227246&coll=portal&dl=ACM > > Derek > > On 2/26/07, John Reppy wrote: > > This week we will start looking at several papers on the topic of > > nested data parallelism. > > The first paper discusses this programming model from a programmer's > > point of view. > > > > Programming Parallel Algorithms > > Guy Blelloch > > CACM, 1996 > > > > It is available in HTML at > > > > http://www.cs.cmu.edu/~scandal/cacm.html > > > > and attached below. The second paper will be the DAMP paper on Data- > > parallel > > Haskell, which focuses more on implementation issues and is also > > attached below. > > > > - John > > > > > > > > _______________________________________________ > > Hyper mailing list > > Hyper@ttic.uchicago.edu > > http://ttic.uchicago.edu/mailman/listinfo/hyper > > > > > > > From jhr at cs.uchicago.edu Mon Feb 26 16:32:22 2007 From: jhr at cs.uchicago.edu (John Reppy) Date: Mon Feb 26 16:40:31 2007 Subject: [Hyper] Nested data-parallel reading In-Reply-To: <7fa251b70702261250o4c986969n6878e59863375af4@mail.gmail.com> References: <7fa251b70702261231t1861fef9lb8f88835dbf019f0@mail.gmail.com> <7fa251b70702261250o4c986969n6878e59863375af4@mail.gmail.com> Message-ID: The room is not available this Thursday, so we'll have to wait until next week. I'll go ahead and switch the reservation of Ry255 to Thursday. - John On Feb 26, 2007, at 2:50 PM, Derek Dreyer wrote: > I assume that, as discussed last week, we will read this for Thursday > (and move the HYPER meeting to Thursday permanently). If there are > any objections to moving to Thursday, then I would suggest we skip > meeting this week and read Blelloch's article for next Tuesday so that > we have time to read it. Meeting tomorrow is too soon. > > Derek > > On 2/26/07, Derek Dreyer wrote: >> The paper is available in PDF here: >> >> http://portal.acm.org/citation.cfm?id=227246&coll=portal&dl=ACM >> >> Derek >> >> On 2/26/07, John Reppy wrote: >> > This week we will start looking at several papers on the topic of >> > nested data parallelism. >> > The first paper discusses this programming model from a >> programmer's >> > point of view. >> > >> > Programming Parallel Algorithms >> > Guy Blelloch >> > CACM, 1996 >> > >> > It is available in HTML at >> > >> > http://www.cs.cmu.edu/~scandal/cacm.html >> > >> > and attached below. The second paper will be the DAMP paper on >> Data- >> > parallel >> > Haskell, which focuses more on implementation issues and is also >> > attached below. >> > >> > - John >> > >> > >> > >> > _______________________________________________ >> > Hyper mailing list >> > Hyper@ttic.uchicago.edu >> > http://ttic.uchicago.edu/mailman/listinfo/hyper >> > >> > >> > >> >