document.write('\x3cmeta http-equiv=\x22x-dns-prefetch-control\x22 content=\x22off\x22/\x3e\x3col class=\x22tumblr_posts\x22\x3e\x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eFree, nicely presented textbooks with good distribution\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cdiv class=\x22posterous_autopost\x22\x3e\x3cp\x3eFree, nicely presented textbooks with good distribution have got quite an appeal.\x3c/p\x3e  \x3cblockquote class=\x22twitter-tweet\x22\x3e\x3cp\x3eBegun it has RT @\x3ca href=\x22https://twitter.com/stevedekorte\x22\x3estevedekorte\x3c/a\x3e: First version of Io Programming Guide available for free on iTunes book store \x3ca href=\x22http://t.co/AovAFyrL\x22 title=\x22http://itunes.apple.com/us/book/io/id497620067?mt=11\x22\x3eitunes.apple.com/us/book/io/id4\u2026\x3c/a\x3e\x3c/p\x3e\u2014 Mac (@wmacgyver) \x3ca href=\x22https://twitter.com/wmacgyver/status/162351174647939073\x22 data-datetime=\x222012-01-26T01:48:31+00:00\x22\x3eJanuary 26, 2012\x3c/a\x3e\x3c/blockquote\x3e      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://justtesting.org/free-textbooks\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/free-textbooks#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eSeminar by John Hughes \x26amp; Simon Peyton Jones @ UNSW\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cdiv class=\x22posterous_autopost\x22\x3e\x3cp\x3eAs part of \x3ca href=\x22http://yowaustralia.com.au/\x22 title=\x22YOW! Conference\x22 target=\x22_blank\x22\x3eYOW! 2011 (the Australian Software Developer Conference)\x3c/a\x3e, \x3ca href=\x22http://research.microsoft.com/en-us/people/simonpj/\x22 title=\x22Simon Peyton Jones\x22 target=\x22_blank\x22\x3eSimon Peyton Jones\x3c/a\x3e and \x3ca href=\x22http://www.cse.chalmers.se/~rjmh/\x22 title=\x22John Hughes\x22 target=\x22_blank\x22\x3eJohn Hughes\x3c/a\x3e will be in Sydney on the 7th and 8th of December 2011. On the evening of the 7th, they will appear at the \x3ca href=\x22http://yownightsydneydec11.eventbrite.com/\x22 title=\x22YOW! Night Sydney\x22 target=\x22_blank\x22\x3eYOW! Night Sydney\x3c/a\x3e.\x3c/p\x3e  \x3cp\x3eOn Thursday (8 December), John and Simon will give two presentations at the School of Computer Science and Engineering (CSE) of the University of New South Wales. The details are as follows.\x3c/p\x3e  \x3cp style=\x22padding-left: 30px;\x22\x3eTime:\u00a0\x3cstrong\x3e8 December 2011, 10AM\x3c/strong\x3e\x3c/p\x3e  \x3cp style=\x22padding-left: 30px;\x22\x3eLocation:\u00a0\x3cstrong\x3eCSE Seminar room (K17_113), Level 1\x3c/strong\x3e, \x3ca href=\x22http://www.cse.unsw.edu.au/contact/map.html\x22 title=\x22UNSW Campus Map\x22 target=\x22_blank\x22\x3eCSE Building (K17)\x3c/a\x3e\x3c/p\x3e    \x3cp\x3e\u00a0\x3c/p\x3e  \x3cp\x3e\x3cspan style=\x22font-size: large;\x22\x3eTalk #1\x3c/span\x3e\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eTitle:\x3c/strong\x3e Accelerating race condition detection through procrastination\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eSpeaker:\x3c/strong\x3e John Hughes (Chalmers University \x26amp; Quviq AB)\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eAbstract\x3c/strong\x3e\x3c/p\x3e  \x3cp\x3eRace conditions are notoriously frustrating to find, and good tools\u00a0can help. The main difficulty is reliably provoking the race\u00a0condition. In previous work we presented a randomising\u00a0scheduler for Erlang that helps with this task.\x3c/p\x3e  \x3cp\x3eIn a language without pervasive shared mutable state, such as Erlang,\u00a0performing scheduling decisions at random uncovers race conditions\u00a0surprisingly well. However, it is not always enough. We describe a\u00a0technique, procrastination, that aims to provoke race\u00a0conditions more often than by random scheduling alone. It works by\u00a0running the program and looking for pairs of events that might\u00a0interfere, such as two message sends to the same process. Having\u00a0found such a pair of events, we re-run the program but try to provoke\u00a0a race condition by reversing the order of the two events.\x3c/p\x3e  \x3cp\x3eWe apply our technique to a piece of industrial Erlang code. Compared\u00a0to random scheduling alone, procrastination allows us to find minimal\u00a0failing test cases more reliably and more quickly.\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eJohn Hughes\x3c/strong\x3e\x3c/p\x3e  \x3cp\x3eJohn Hughes has worked with functional programming since 1980,\u00a0was one of the designers of Haskell, and has been heavily involved\u00a0with Erlang in recent years. In 2000 he and Koen Claessen published\u00a0the first version of QuickCheck, a software testing tool which recently\u00a0won the ACM SIGPLAN award for Most Influential Paper from ICFP\u00a0in that year. He has focussed more and more on testing since then,\u00a0co-founding Quviq AB in 2006 to market a commercial version of\u00a0QuickCheck. He is currently both a Professor at Chalmers University,\u00a0Sweden, and CEO of Quviq AB.\x3c/p\x3e    \x3cp\x3e\x3cspan style=\x22font-size: large;\x22\x3eTalk #2\x3c/span\x3e\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eTitle:\x3c/strong\x3e Termination Combinators Forever\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eSpeaker:\x3c/strong\x3e Simon Peyton Jones (Microsoft Research)\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eAbstract\x3c/strong\x3e\x3c/p\x3e  \x3cp\x3eNobody wants their compiler, or theorem prover, to go into an infinite loop,\u00a0but making sure that never happens usually means applying some over-conservative\u00a0heuristic like \u201cnever unfold a recursive function\u201d. Approaches like that don\u2019t\u00a0work at all when you are doing partial evaluation or supercompilation, which\u00a0fundamentally depend on inlining recursive functions.\u00a0\x3c/p\x3e  \x3cp\x3eWhat we need is an online termination test. That is easier said than done; it\u00a0is easy to make a mistake. In this talk I\u2019ll describe a new combinator library\u00a0that lets you build complex termination tests by combining simpler ones, while\u00a0guaranteeing that that the result really is a termination test. The library\u00a0elegantly encapsulates some clever mathematical ideas on well-quasi orders the\u00a0cunning details are hidden from the customer.\x3c/p\x3e  \x3cp\x3eI\u2019ll use Haskell as the language of exposition, but you don\u2019t need to be a\u00a0Haskell guru to understand it, and the library would work equally well in other\u00a0languages.\x3c/p\x3e  \x3cp\x3e\x3cstrong\x3eSimon Peyton Jones\x3c/strong\x3e\x3c/p\x3e  \x3cp\x3eSimon Peyton Jones, MA, MBCS, CEng, graduated from Trinity College Cambridge in\u00a01980. After two years in industry, he spent seven years as a lecturer at\u00a0University College London, and nine years as a professor at Glasgow University,\u00a0before moving to Microsoft Research (Cambridge) in 1998.\x3c/p\x3e  \x3cp\x3eHis main research interest is in functional programming languages, their\u00a0implementation, and their application. He has led a succession of research\u00a0projects focused around the design and implementation of production-quality\u00a0functional-language systems for both uniprocessors and parallel machines. He was\u00a0a key contributor to the design of the now-standard functional language Haskell,\u00a0and is the lead designer of the widely-used Glasgow Haskell Compiler (GHC). He\u00a0has written two textbooks about the implementation of functional languages.\x3c/p\x3e  \x3cp\x3eMore generally, he is interested in language design, rich type systems, software\u00a0component architectures, compiler technology, code generation, runtime systems,\u00a0virtual machines, and garbage collection. He is particularly motivated by direct\u00a0use of principled theory to practical language design and implementation \u2014 that\x26#8217;s one reason he loves functional programming so much.\x3c/p\x3e        \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://pls.posterous.com/seminar-by-john-hughes-simon-peyton-jones-uns\x22\x3ePLS @ UNSW\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://pls.posterous.com/seminar-by-john-hughes-simon-peyton-jones-uns#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eVideo and slides of \x26#8220;Data Parallelism in Haskell\x26#8221; @ Brisbane FP Group\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3ciframe frameborder=\x220\x22 width=\x220\x22 height=\x220\x22\x3e\x3c/iframe\x3e\x3c/p\x3e\x0a\x3cp\x3e\x3ca title=\x22Data Parallelism in Haskell\x22 href=\x22http://vimeo.com/28477220\x22\x3eVideo @ vimeo\x3c/a\x3e\x3c/p\x3e\x0a\x3cp\x3eThe\u00a0\x3ca href=\x22http://www.bfpg.org/\x22\x3eBrisbane FP Group (BFPG)\x3c/a\x3e\u00a0kindly invited me to give a talk about our work on data parallel programming in Haskell. The talk motivates the use of functional programming for parallel, and in particular, data parallel programming and explains the difference between regular and irregular (or nested) data parallelism. It also discusses the\u00a0\x3ca href=\x22http://hackage.haskell.org/package/repa\x22\x3eRepa\x3c/a\x3e\u00a0library (regular data parallelism for multicore CPUs), the embedded language\u00a0\x3ca href=\x22https://github.com/mchakravarty/accelerate\x22\x3eAccelerate\x3c/a\x3e\u00a0(regular data parallelism for GPUs), and Data Parallel Haskell (nested data parallelism for multicore CPUs).\u00a0 The slides of the presentation are available in two formats:\u00a0\x3ca href=\x22http://public.iwork.com/document/?d=dp-in-haskell_BFPG_Sep_2011.key\x26amp;a=p179149432\x22\x3eHTML5 slideshow\x3c/a\x3e\u00a0and\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/papers/dp-in-haskell.pdf\x22\x3ePDF\x3c/a\x3e. Thank you to everybody who attended. Special thanks go to OJ Reeves and Tony Morris for organising everything, to Rob Manthey for producing the video, to Mincom for the venue, and to\u00a0\x3ca href=\x22http://functional.io/\x22\x3eFunctional IO\x3c/a\x3e\u00a0for the sponsorship.\x3c/p\x3e\x0a\x3cp\x3e\x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e from \x3ca href=\x22http://justtesting.org/video-and-slides-of-data-parallelism-in-haske\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/video-and-slides-of-data-parallelism-in-haske#comment\x22\x3e\x3cspan\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eData Parallel Haskell and Repa for GHC 7.2.1\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3eAs an add-on to the just released\u00a0\x3ca href=\x22http://www.mail-archive.com/glasgow-haskell-users@haskell.org/msg20202.html\x22\x3eGHC 7.2.1\x3c/a\x3e, Ben just uploaded the\u00a0\x3ca href=\x22http://www.haskell.org/haskellwiki/GHC/Data_Parallel_Haskell\x22\x3eData Parallel Haskell\x3c/a\x3e\u00a0packages (version 0.5) to Hackage. This version is still largely a technology preview and not a production-ready system. Nevertheless, it is significantly more robust than previous versions, especially for programs with a statically fixed depth of nesting of parallelism. For further information on how to install and use Data Parallel Haskell (DPH), see the\u00a0\x3ca href=\x22http://www.haskell.org/haskellwiki/GHC/Data_Parallel_Haskell\x22\x3eDPH documentation.\x3c/a\x3e\x3cdiv\x3eBen simultaneously released the companion library\u00a0\x3ca href=\x22http://hackage.haskell.org/package/repa\x22\x3eRepa\x3c/a\x3e\u00a0(version 2.1.1.6) for use with GHC 7.2.1. In contrast to the DPH libraries, which enable \x3ci\x3enested data parallelism\x3c/i\x3e, Repa implements parallel, shape-polymorphic, \x3ci\x3eregular\x3c/i\x3e\u00a0multi-dimensional arrays. \u00a0See\u00a0\x3ca href=\x22http://pls.posterous.com/real-time-edge-detection-with-the-latest-rele\x22\x3ethis previous post\x3c/a\x3e\u00a0for some more details.\x3c/div\x3e      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://pls.posterous.com/data-parallel-haskell-and-repa-for-ghc-721\x22\x3ePLS @ UNSW\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://pls.posterous.com/data-parallel-haskell-and-repa-for-ghc-721#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eData.Array.Accelerate now on GitHub\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3ePrompted by GHC\x26#8217;s move to Git and the unreliability of the \x3ca href=\x22http://community.haskell.org\x22\x3ecommunity.haskell.org\x3c/a\x3e infrastructure, I decided to move\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/project/accelerate/\x22\x3eData.Array.Accelerate\x3c/a\x3e\u00a0to GitHub:\u00a0\x3ca href=\x22https://github.com/mchakravarty/accelerate\x22\x3emchakravarty/accelerate\x3c/a\x3e. The GitHub repository is now the main public repository for the project. I will also move the tickets of the old Trac bug tracker over to GitHub Issues.      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://justtesting.org/dataarrayaccelerate-now-on-github\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/dataarrayaccelerate-now-on-github#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eReal-time edge detection with the latest release of the parallel array library Repa\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3eWe just\u00a0\x3ca href=\x22http://hackage.haskell.org/package/repa\x22\x3ereleased version 2.0.0.1\x3c/a\x3e\u00a0of the Repa array library for Haskell, which includes Ben\x26#8217;s recent work on high-performance, parallel stencil computations. The work on stencil computations is described in detail in a draft paper entitled\u00a0\x3cbr/\x3e\x3ca href=\x22http://www.cse.unsw.edu.au/~benl/papers/stencil/stencil-icfp2011-sub.pdf\x22\x3eEfficient Parallel Stencil Convolution in Haskell\x3c/a\x3e. Be sure to check out\u00a0\x3ca href=\x22http://code.ouroborus.net/beholder/video/Edge480.mov\x22\x3eBen\x26#8217;s screencast\x3c/a\x3e\u00a0of a real-time edge detection application, written in Objective-C and Haskell, using the new Repa library. For more details, see\u00a0\x3ca href=\x22http://disciple-devel.blogspot.com/2011/03/real-time-edge-detection-in-haskell.html\x22\x3eBen\x26#8217;s blog post.\x3c/a\x3e      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://pls.posterous.com/real-time-edge-detection-with-the-latest-rele\x22\x3ePLS @ UNSW\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://pls.posterous.com/real-time-edge-detection-with-the-latest-rele#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eSimon Peyton Jones: parallel = functional\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3eSimon Peyton Jones recently gave a talk on parallel programming in Haskell at\u00a0\x3ca href=\x22http://skillsmatter.com/event/scala/functionalpx-2011\x22\x3eFunctional Programming eXchange 2011\x3c/a\x3e. It provides a comprehensive overview of the state of the art of parallel programming in Haskell. It covers Software Transactional Memory, Erlang-style communicating processes, parallel strategies as well as our Data Parallel Haskell, Repa, and Accelerate projects.\u00a0\x3ca href=\x22http://skillsmatter.com/podcast/scala/talk-by-haskell-expert-simon-peyton-jones/js-1434\x22\x3eWatch the video.\x3c/a\x3e      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://pls.posterous.com/simon-peyton-jones-parallel-functional\x22\x3ePLS @ UNSW\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://pls.posterous.com/simon-peyton-jones-parallel-functional#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eMoving mail archives to MobileMe (or other Sun Java System Messaging Servers) (tag: ipad, mac, mobileme, mail, imap)\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cdiv class=\x22posterous_autopost\x22\x3eI recently consolidated all my email on MobileMe (for easy access from both my Mac and iPad). \u00a0However, I ran into a problem with the MobileMe IMAP server when I tried to move my old mail archives to the MobileMe IMAP server. \u00a0MobileMe uses the\u00a0\x3ca href=\x22http://wikis.sun.com/display/CommSuite6/Messaging+Server+7+Documentation\x22\x3eSun Java System Messaging Server\x3c/a\x3e\u00a0(which has since become the\u00a0\x3ca href=\x22http://en.wikipedia.org/wiki/Oracle_Communications_Messaging_Exchange_Server\x22\x3eOracle Communications Messaging Exchange Server\x3c/a\x3e). \u00a0It turns out that this IMAP server (like some others as well) is rather strict about the format of mail headers; hence, uploading old mail archives usually aborts with the following error message: \x26#8220;The IMAP command \u201cAPPEND\u201d (to \u2026) failed with server error: Message contains invalid header\x26#8221;.\x3cdiv\x3eQuerying the Internets and some experimentation revealed that the problem are lines in email message headers that start with \x26#8220;From \x26#8220;, \x26#8220;\x26gt;From \x26#8220;, and \x26#8220;\x26#187;From \x26#8220;. \u00a0These are not proper message headers fields (whose names need to be delimited by a colon), but remnants from storage of these mail messages in the\u00a0\x3ca href=\x22http://en.wikipedia.org/wiki/Mbox\x22\x3embox format\x3c/a\x3e. \u00a0The remedy is to delete or edit these invalid lines in all affected message headers.\x3c/div\x3e\x3cp\x3e\x3c/p\x3e\x3cdiv\x3eApple Mail stores mail messages in \x26#8220;$USER/Library/Mail/Mailboxes/\x26#8221;. \u00a0Each mailbox directory (suffix \x26#8220;.mbox\x26#8221;) contains a directory \x26#8220;Messages\x26#8221; with one file per message. \u00a0These files can easily be modified using your favourite command line tools that support matching of regular expressions or by loading them, en masse, into TextMate for project-wide search and replace. \u00a0This is somewhat naughty as the files containing individual messages contain the message header and text as well as a property list (plist), used by Mail.app to store a few attributes. \u00a0The start of that property list in the text file is marked by a character count in the first line of each message. \u00a0By eliminating or modifying the offending lines, you invalidate that character count. \u00a0However, Mail.app seems to cope just fine with those slightly malformed messages. \u00a0After fixing the headers, all my messages uploaded without problems (and no attributes seemed to get lost).\x3c/div\x3e      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://justtesting.org/moving-mail-archives-to-mobileme-or-other-sun\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/moving-mail-archives-to-mobileme-or-other-sun#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eFinal version of the Accelerate paper (GPGPU in Haskell)\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3eWe will present our paper\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/papers/CKLM+11.html\x22\x3eAccelerating Haskell Array Codes with Multicore GPUs\x3c/a\x3e\u00a0at\u00a0\x3ca href=\x22http://damp2011.cs.uchicago.edu/\x22\x3eACM SIGLAN\u00a0Declarative Aspects of Multicore Programming (DAMP 2011)\x3c/a\x3e, which is co-located with POPL\x26#8217;11 in Austin, TX, in January. The final version of our paper is\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/papers/CKLM+11.html\x22\x3enow available\x3c/a\x3e, and we plan to soon release a significantly improved version of the\u00a0\x3ca href=\x22http://hackage.haskell.org/package/accelerate\x22\x3eAccelerate library\x3c/a\x3e\u00a0(matching the API used in the paper), which enables high-level GPGPU programming in Haskell based on NVIDIA\x26#8217;s CUDA environment.      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://justtesting.org/final-version-of-the-accelerate-paper-gpgpu-i\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/final-version-of-the-accelerate-paper-gpgpu-i#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a        \x0a            \x3cli class=\x22tumblr_post tumblr_text_post\x22\x3e\x0a                \x0a                    \x3cdiv class=\x22tumblr_title\x22\x3eFinal version of the Singleton paper\x3c/div\x3e\x0a                \x0a                \x0a                \x3cdiv class=\x22tumblr_body\x22\x3e\x0a                    \x3cp\x3e\x3cdiv class=\x22posterous_autopost\x22\x3eOur paper\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/papers/WC11.html\x22\x3eSingleton: A General-Purpose Dependently-Typed Assembly Language\x3c/a\x3e\u00a0will be presented at the\u00a0\x3ca href=\x22http://www.mpi-sws.org/~dreyer/tldi2011/\x22\x3eACM\u00a0SIGPLAN\u00a0Workshop on\x3cbr/\x3eTypes in Language Design and Implementation (TLDI\x26#8217;11)\x3c/a\x3e\u00a0co-located with POPL\x26#8217;11 in\u00a0Austin, TX,\u00a0in January. \u00a0The final version of the paper is now\u00a0\x3ca href=\x22http://www.cse.unsw.edu.au/~chak/papers/singleton.pdf\x22\x3eavailable\x3c/a\x3e.      \x3cp style=\x22font-size: 10px;\x22\x3e  \x3ca href=\x22http://posterous.com\x22\x3ePosted via email\x3c/a\x3e   from \x3ca href=\x22http://justtesting.org/final-version-of-the-singleton-paper\x22\x3eJust Testing\x3c/a\x3e\u00a0|\u00a0\x3ca href=\x22http://justtesting.org/final-version-of-the-singleton-paper#comment\x22\x3e\x3cspan style=\x22font-size: 11px\x22\x3eComment\x26#160;\u00bb\x3c/span\x3e\x3c/a\x3e  \x3c/p\x3e  \x3c/div\x3e\x3c/p\x3e\x0a                \x3c/div\x3e\x0a            \x3c/li\x3e\x0a        \x0a\x0a        \x0a\x0a        \x0a\x0a        \x0a        \x0a        \x0a        \x0a        \x0a\x0a                        \x0a    \x0a\x3c/ol\x3e');