Squeak-dev October 2001

squeak-dev@lists.squeakfoundation.org

191 participants
600 discussions

Re: Lots of concurrency
by Richard A. O'Keefe 29 Oct '01

29 Oct '01

Andres Valloud <sqrmax(a)prodigy.net> wrote: Descartes said "I think, therefore I am". Maybe a more illuminating thing to say in this context would be "I experience self-consciousness, therefore I am". If we go down that road, then many things seem more clear. Our self-consciousness "processor" runs threads. Threads tap on shared brain resources, they also have a dedicated storage space we may refer to as its state. The running thread can be interrupted by other threads. Any interrupting thread may become the running thread. Threads may call other threads, return, terminate, etc. Keeping a thread stack requires concentration. I suggest that this is not a useful way to think about thinking. Some years ago I noticed that if the instruments were sufficiently distinct in timbre or pitch, I could follow more than one musical part at once, with no _experienced_ switching. I could always "attend" to three parts at once, could "attend" to four parts at once more often than not, and on one occasion only managed to track five musical parts. I want to emphasise that there were no experienced* attention switches. I know what an attention switch feels like, and these days when listening to four parts experience them quite often. I can still manage three parts, though. At one and the same time, I am (as far as I can tell) _continuously_ attending to several musical parts, and there is another aspect of me that notices that I am doing this. The "attention points" don't feel as if they are in the same place in my head, although I am confident that this attention-point-location thing is an illusion. If I start thinking about anything, my capacity drops, and my musical attention starts switching. I don't believe that I am unique. Another data point. While I am typing this and thinking about what to say, I can whistle, hum, or sing, as long as I don't sign the words. When I try to sing the words, my typing stops. I suspect, therefore, that the thing which is single-threaded is the speech centre, and that it is _verbal_ thinking at most which is single-threaded, NOT consciousness. The relevance of this to things like Squeak and EToys is that programming by direct manipulation may have different consequences for "thought" than programming by typing words in a code pane, and that the metaphors we give children (about TELLING little men what to do) may be the source of sequential thinking, not any intrinsic limitation in the children themselves.

2 1

[bioSqueak] [Open-Bioinformatics-Foundation] All Project Open-Bio Newsletter
by Bruce ONeel 29 Oct '01

29 Oct '01

This might interest some of you... ---------- Forwarded ---------- Date: Wed, 10 Oct 2001 15:20:16 -0400 (EDT) From: Chris Dagdigian <dag(a)sonsorol.org> Subject: [Open-Bioinformatics-Foundation] All Project Open-Bio Newsletter (long) To: <open-bioinformatics-foundation(a)open-bio.org> X-x-sender: <dag(a)localhost.localdomain> X-beenthere: open-bioinformatics-foundation(a)open-bio.org X-mailman-version: 2.0.1 List-help: <mailto:open-bioinformatics-foundation-request@open-bio.org?subject=help> List-post: <mailto:open-bioinformatics-foundation@open-bio.org> List-subscribe: <http://open-bio.org/mailman/listinfo/open-bioinformatics-foundation>, <mailto:open-bioinformatics-foundation-request@open-bio.org?subject=subscribe> List-id: One-time list for all-project broadcast messages <open-bioinformatics-foundation.open-bio.org> List-unsubscribe: <http://open-bio.org/mailman/listinfo/open-bioinformatics-foundation>, <mailto:open-bioinformatics-foundation-request@open-bio.org?subject=unsubscribe> List-archive: <http://open-bio.org/pipermail/open-bioinformatics-foundation/> This email is being sent to every person who is subscribed to one of the discussion or announce lists that we host. Rather than cross post to 31 individual mailing lists we merged all the subscribers into a unique list. This helps to keep our active volunteers from getting many copies of the same message. =========================================================================== (O|B|F) Open Bioinformatics Foundation N E W S L E T T E R October 2001 http://open-bio.org biopython.org, biojava.org, bioperl.org bioXML.org, bioCORBA.org, bioDAS.org, biomoby.org, etc. =========================================================================== Introduction There has been very significant progress and change within our project(s) and organizational ranks lately; most of these changes not been immediately obvious or visible on our web sites or mailing lists. This email message is our first attempt at what will hopefully become a regular update on the 'big picture' status of our projects and efforts. If you have comments, questions or concerns about anything in this newsletter you can email the Open-Bio board directly at <board(a)open-bio.org> Summary of topics in today's message: o Organizational Status - Our cross-project name is now the "Open Bioinformatics Foundation" - O|B|F is now a non-profit corporation - 2001/2002 Board of Directors announced - Legal services provided pro bono by HellerEhrman o Organizational Financial Summary - Current funds - BOSC'2001 profit o Server and Connectivity Update o Current & New project briefs - New efforts: bioMOBY & bioSOAP o Upcoming Events - ORA Bioinformatics Technology Conference + BOF leaders needed ASAP - Open-Bio Hackathon(s) + Phoenix, Arizona + Cape Town, South Africa o Call for Volunteers ------------------------------------------------------------------------- ORGANIZATIONAL STATUS & NEW LEGAL REPRESENTATION ------------------------------------------------------------------------- Over the past few years an informal group of volunteers and open-bio project leaders has self-formed to handle issue that span all of our current projects. Many of these new needs came about as we began to 'own' assets such as servers and domain names and others happened once we began organizing bootcamps and conferences and found ourselves having to sign binding legal and financial agreements. This group has also handled all the behind the scenes work necessary to keep our servers and Internet connectivity running smoothly. It has become obvious over the past year that this "umbrella" group while functional could benefit greatly from a more formal organizational structure. Long talked about it was finally decided at BOSC'2001 in Copenhagen that we should take the plunge and incorporate the group as a formal not-for-profit entity. The goals of the new organization are the same as before: providing administrative, financial and technical support to our ongoing and future projects. The new entity will be called the "Open Bioinformatics Foundation" The current Directors for 2001/2002 are as follows: - Ewan Birney (European Bioinformatics Institute) - Steven E. Brenner (University of California, Berkeley) - Andrew Dalke (Dalke Scientific Software, LLC ) - Chris Dagdigian (Blackstone Computing Inc.) - Hilmar Lapp (Novartis Research Foundation) The Directors have chosen corporate Officers for the following positions: - Ewan Birney, President & Chief Executive Officer - Chris Dagdigian, Treasurer & Chief Financial Officer - Andrew Dalke, Secretary A rented mailbox serves as our official corporate address: Open Bioinformatics Foundation 411A Highland Avenue #318 Davis Square Somerville, MA 02144 Phone/Fax 617-250-0000 x4327 As of October 4th 2001 the Open Bioinformatics is a not-for-profit company incorporated in the state of Delaware. More info and links to our meeting minutes and organizational bylaws will be forthcoming. New Legal Representation for the organization As part of the ongoing attempts to get ourselves organized we are very pleased to announce that we now have top-notch legal assistance being provided pro-bono by the law firm HellerEhrman (http://www.hewm.com) In particular we'd like to acknowledge the assistance and guidance of Dan Appelman who co-chairs the IT National Practice Group at HellerEhrman. His attorney bio can be read online at http://www.hewm.com/attorneys/attorneyBio.asp?attorneyID=341 ------------------------------------------------------------------------- FINANCES ------------------------------------------------------------------------- Our current bank account is the same one we have been using for several years-- a "Small Business" account registered to "The BioPerl Project" held at a bank in Massachusetts, USA. After the corporation is fully formed it is likely that will close this account and open a new "Open Bioinformatics Foundation" account. Our current bank balance is approximately $7000 USD of which most is earmarked to pay for BOSC'2000 facility expenses that UCSD is very late in billing us for. This means that practically speaking we have less than $1000 USD free cash at the moment. Most if not all of that remaining money will be used to pay filing fees and expenses associated with incorporating the non-profit company. Sun Microsystems had provided significant financial support to offset BOSC'2001 expenses. We have invoiced them for the full amount but won't consider it 'real money' until we receive the funds. Future expenses that we foresee: o Misc. hardware & gear needed for racking our new server systems o Getting a Sun hardware support contract for the donated systems o Purchase/renewal of domain names o Supporting hackfests & misc. activities Needless to say cash or hardware donations are welcome. BOSC'2001 (Copenhagen) Financial Summary Despite a very successful conference in Copenhagen we had some significant expenses caused mainly by the requirement that we use the designated ISMB conference company to handle registration and AV rental support. Our goal in general with BOSC meetings is to make the registration fee as low as possible while trying to ensure that we don't actually lose any money. We have made a small profit at each of the last 2 conferences. BOSC'2001 (Denmark) Conference financial breakdown: Total number of attendees: 163 Our fixed costs per attendee were 880 DKK per person Income received: Academic: 57 @ 1100 DKK = 62700.00 DKK Corporate: 68 @ 1400 DKK = 95200.00 DKK Student: 38 @ 880 DKK = 33440.00 DKK Total income: 191340.00 DKK Expenses: Meeting expenses (fixed) 158933.00 DKK Extra AV costs 19582.00 DKK Poster stands 2640.00 DKK Extra flipcharts 594.00 DKK Total expenses: 181749.00 DKK BOSC'2001 PROFIT: 9591.00 DKK ($1,177.00 US Dollars) BOSC Pictures available at http://open-bio.org/bosc2001/bosc2001_pics/ ------------------------------------------------------------------------- SERVER & CONNECTIVITY UPDATE ------------------------------------------------------------------------- Our upstream provider of donated internet bandwidth is upgrading its internet connection through October and November. There may be downtime or sporadic connectivity outages as this effort progresses. If we need to change the IP addresses of our servers the downtime may last 8-24 hours as the new domain info propagates outward. Thanks again to Genetics Institute / Wyeth Ayerst Research we now have secure space in which to begin unpacking and building our new server hardware. A short list of the hardware we have available is as follows: o 3 Sun Netra T1000 high density rackmount servers o 1 Sun Netra A1000 UltraSCSI RAID array o 1 Cobalt Raq 4 rackmount server appliance o 1 VALinux 1220 high density PentiumIII rackmount server Our current plans are to split the multiple servers out according to task: 1) Web, email listserv, DNS & FTP services 2) Core project(s) server with RAID for our source code and developers 3) anonymous CVS front-end and nightly build system A firewall/IDS system is being worked on as a separate project. We will be building, integrating and rolling out these systems in stages over the next several months. Expect to see a few more announcements and solicitations for volunteer assistance as things get under way. We need people with Solaris Admin skills to help us build, tune and secure the Netra servers. See the section below on "Volunteers" for more information or contact Chris Dagdigian directly at dag(a)sonsorol.org. New hardware pictures are online. We will be updating this URL as the build out of our new server systems continues: ** http://open-bio.org/Hardware-pics/ ** Some website statistics: Apache Server Statistics as of 2 October 2001 Server uptime: 16 days 15 hours 11 minutes 9 seconds Total accesses: 146,749 - Total Traffic: 2.4 GB Bioperl.org Website stats 1 Year Average: 98,172 hits/month 59,925 pageviews/month Sep 2001 : 62,125 hits, 874,361 KB transferred, 86 hits/hour avg Biojava.org Website stats 1 Year Average: 57,475 hits/month 34,389 pageviews/month Sep 2001 : 66,418 hits, 2,291,842 KB transferred, 92 hits/hour avg ------------------------------------------------------------------------- INCOMPLETE PROJECT BRIEFS & ANNOUNCEMENT OF NEW PROJECTS ------------------------------------------------------------------------- o Biopython Biopython is rapidly approaching a 1.0 release with a maturing sequence model, support for pathways, a new parsing framework (Martel), and algorithms for sequence computation and alignment. Currently, we are adding support for more file formats and implementing more algorithms. As always, volunteers are welcome. In addition to coders, we also need people to work on testing, documentation, and web development. o Biojava Since the 1.1 release earlier this year, the emphasis of BioJava development has been on improved connectivity (DAS, Ensembl), and better flatfile parsers (EMBL, Genbank, Blast). In the future we hope to see more support for ontologies, and a more general query mechanism. There is also a group interested in developing an object model for expression data. All contributions are welcome, whether documentation, code, or suggestions. Additional tutorials are always extremely welcome! o Biocorba The BioCORBA project has seen increased activity after building numerous prototype and production servers. Previously, two standards for biological objects in CORBA existed from the BioCORBA project and the BSA and BSANE proposals from the LSR group of the OMG (http://lsr.omg.org). After the meetings held at ISMB 2001 in Copenhagen, Denmark (dubbed the "Tivoli Meetings") a compatible proposal was agreed upon which combined these two standards. Work is under way in the Perl and Python camps to implement these standards. There is a need for Java volunteers to implement a BioJava <-> BioCorba bridge in the months to come. The end result of these standards will allow seamless launching of analysis applications in perl,python,and java (even C if someone wants to implement the client side). Connections to sequence (EMBL, NCBI) and annotation sources (Ensembl, Flybase, WormBase) will allow developers to integrate data sources with analysis systems. This will further simplify the establishment of pipelines for both small laboratories and large institutions which wish to rely on open-bio toolkits for their informatics needs. Tutorials and Full documentation are being generated as well as skeleton programs to serve as examples. Opportunities for volunteers exists in every aspect of the project and from evaluating the BioCORBA/BSANE standard to writing client/server software to helping produce documentation and testing applications for portability and ease of use. o Bioperl The Bioperl project continues to expand and address more needs of biological researchers. New developers and contributors to the mailing lists have furthered the scope of the project and seek to address new areas of data from microarray, phylogenetics, bibliographic, and annotation sources as well as integrate with more external applications. Recent work has produced modules suitable for retrieving sequences from online sources such as swissprot and EMBL. One can also submit analysis jobs to online analysis queues such as NCBI's blast queue and soon we will have build access to EBI applab analysis through the NOVELLA CORBA interfaces. New work is focusing on standardizing input and output methods to these application servers. Additional resources for local DAS servers in bioperl have been submitted and will be part of the next major release of bioperl. We expect the 1.0 release to be completed by the end of 2001 and will continue to be a stable platform for bioinformatics software development in perl. The 0.7.2 stable release is expected to be released in mid November and will correct a number of small bugs in the 0.7.1 release. Developer releases (0.9.x unstable series) will continue to be released until the 1.0 release and serve as snapshops of working code in the release that passes all bioperl tests. o BioXML o BioDAS Two new efforts have been initiated recently o BioMOBY (http://biomoby.org) ** site online ** BioMOBY is an international group of biological data hosts, biological data service providers, and coders whose aim is to achieve a maximum amount of data interoperability between host institutions. The website provides an online resource for modules, scripts, and schema for developers of MOBY-related software. CVS access will be available shortly. BioMOBY project admin: Mark Wilkinson <mwilkinson(a)gene.pbi.nrc.ca> o BioSOAP (http://biosoap.org) ** future site ** biosoap.org will aim to develop a bridge system between the core objects of bioperl and biojava as well as between the current perl-based Ensembl and the upcoming java port of it. It will hopefully allow such things as perl scripts running with java objects, as well as over-the-net object oriented programming. It is in its infancy, so any support, advice and suggestions will be most appreciated. BioSOAP project admin: Elia Stupka <elia(a)ebi.ac.uk> o biostandards.org We've had this domain name for a while and have done nothing with it. Suggestions welcome. ------------------------------------------------------------------------- UPCOMING EVENTS ------------------------------------------------------------------------- 2-Part Open Bioinformatics Hackathon in USA and South Africa ! This is an early sneak-peek announcement. Details will follow from the actual organizers. O'Reilly & Associates (http://www.ora.com) and Electric Genetics (http://www.egenetics.com) are going to jointly hold two open-bio hackathons. The hackfest will start with a three-day event as part of the O'Reilly Bioinformatics Technology Conference in Arizona in late January and continue with a seven-day session in Cape Town, South Africa scheduled for late February. The event is invitation only and attendees will have all travel and accomodation expenses paid. This is very exiting news, more details will be announced. Open-Bio at the O'Reilly Bioinformatics Technology Conference, Jan 2002 Conference site: http://conferences.oreilly.com/biocon/ It appears that many of our project admins and developers are going to be attending the ORA conference in January. Ewan Birney is giving a keynote address and several Open Bio people are giving talks and/or tutorials at the conference. In addition to the tutorial sessions, the OBF will also have a conference booth in the exhibit hall and we have been invited by the conference staff to organize and host informal Bird Of a Feather ("BOF") sessions as necessary. We have already committed to hosting the following BOF sessions at the conference: o Bioperl developers o new Bioperl users o Biopython users & developers HOST YOUR OWN BOF! If you are interested in hosting/moderating a gathering of like-minded individuals at the conference please respond to Andrew Dalke <dalke(a)dalkescientific.com>. We need to know (a) who you are and (b) what BOF topic you propose to host. ------------------------------------------------------------------------- CALL FOR VOLUNTEERS ------------------------------------------------------------------------- We always welcome volunteers and are soliciting people right now for the following roles and projects: * Volunteer Coordinator This person or group would be responsible for monitoring a new email address we are going to set up called "volunteer(a)open-bio.org". The coordinators will screen & manage volunteer offers, match them to current projects/needs and otherwise help with orientation and introductions. Time commitment: Minimal How to sign up : email board(a)open-bio.org * Webteam (architect & deploy a new website for us) The problem is simple. Our web presence sucks. We need something better. The task of the webteam is to come up with a plan and follow it through. Time commitment: depends; may be significant in the early stages How to sign up: email volunteer(a)open-bio.org * Mailteam We use the GNU Mailman system for our mailing lists. It is a nice piece of software that can be almost entirely administrated via a web interface. Once the lists are created they are largely self-operating. We occasionally need an administrator to respond to user question, manually unsubscribe the clueless and act as moderators when our spam mail filters quarantine suspect messages. We are looking to set up a small group of people who monitor and moderate the dozens of mailing lists we have currently running. Time commitment: minimal How to sign up: email volunteer(a)open-bio.org * SolarisGurus A significant amount of our new server hardware will run Solaris. We need people who are available to answer configuration questions, idiot check changes and help with securing each box. Time commitment: depends; may be significant in the early stages How to sign up: email volunteer(a)open-bio.org * CambridgeTeam All of our existing and future server hardware is located in Cambridge, Massachusetts (subway accessible via the Red Line!). There may come a time where we need physical bodies to help rack servers or possibly transport the systems to new hosting facilities. We are attempting to get a sense of how many people we have local to the area that may be available if needed. Time commitment: probably zero How to sign up: email Chris Dagdigian <dag(a)sonsorol.org> _______________________________________________ Open-Bioinformatics-Foundation mailing list Open-Bioinformatics-Foundation(a)open-bio.org http://open-bio.org/mailman/listinfo/open-bioinformatics-foundation -------------------------------

1 0

Re: Lots of concurrency
by Russell Allen 29 Oct '01

29 Oct '01

Hi, I like the way in which the squeak mailing list attempts to merge concepts from outside the CompSci universe into discussions - recently I have been intrigued by what other people are saying about how they perceive music. For me, I can listen to music in the way which people are suggesting (as a parallel series of notes - ie horizontally) or vertically as a succession of sounds/chords, or I think, in whole number of other ways less amenable to computing analogies ;). This, though, is just my experience. I could, for what it is worth, tell the list about how I perceive music, and how this changes when, for example, I am singing a solo. I wouldn't want to tell the list (without some serious research) what all soloists think/experience when singing. I have a strong suspicion that this retisence should be carried over to other sweeping generalisations: I know both male and female musicians who can 'multitask' better than I can. Whether there is a fundamental difference in brain make-up is still undecided; and "Men are from Mars and Women are from Venus" is hardly a groundbreaking piece of either sociology or biology. I am also faintly amused by the thought that I (a male) am commenting on Justin (a male), who is quoting a book by John (a male) on the subject of women's mentalities... Are there any women on this list? And if so, why not? What good is programming for children of all ages if half the children are uninterested or excluded? Russell "Justin Walsh" <jwalsh(a)bigpond.net.au> wrote: > It is true John. I have my own proof but, > John Grey the Author of the "Men are from Mars and Women are from Venus" (ex > budist monk of 28 years I believe) describes it very well. > > ----- Original Message ----- > From: "John Hinsley" <jhinsley(a)telinco.co.uk> > To: <squeak-dev(a)lists.squeakfoundation.org> > Sent: Monday, October 29, 2001 6:36 PM > Subject: Re: Lots of concurrency > > > > As a footnote, it would seem that women have brains which multi-task > > much better than men's. > > > > Any evidence for enhanced parallelism in women, or is this just a crude > > OS characteristic? > > > > Cheers > > > > John > > > > -- > > If you don't care about your data, like file systems which automagically > > destroy themselves and have money to burn on 3rd party tools to keep > > your > > system staggering on, Microsoft (tm) have the Operating System for you. > > > >

1 0

Debian source package
by Serge Stinckwich 29 Oct '01

29 Oct '01

Hi all, i'm looking for the debian package source for Squeak 3.0. Debian binary packages (made by Markus Denker) seems to be available, but i can't find the source one. Where are they ? -- Serge Stinckwich -< ) http://multiagent.com/ CNRS UMR 6072>GREYC>I3>SMILE /~\ http://squeak.org/ Université de Caen>IUT de Caen>Campus 3>Dept Info (/ | http://smalltalk.org/ http://www.iutc3.unicaen.fr/~stincs/ _|_/ http://zope.org/ JID : Serge.Stinckwich(a)jabber.org http://debian.org/

4 3

[ANNC] Project Swiki (Beta)
by Jochen F. Rick 29 Oct '01

29 Oct '01

For your consideration... http://manatee.cc.gatech.edu:8080/beta This site will only be up for a limited time. Peace and Luck! Je77

2 1

Re: Corrupted sources?!
by Bob Arning 29 Oct '01

29 Oct '01

The problem is probably in the pointer(s) from the image to the .changes file, not the .sources file. Cheers, Bob On Mon, 29 Oct 2001 06:50:13 -0500 Kevin Fisher <kgf(a)golden.net> wrote: >On Sun, Oct 28, 2001 at 07:14:31PM -0500, David T. Lewis wrote: >[snip] >> >> I don't see the problem on Squeak3.1beta latest update: #4411. >> >> $ ls -l *sources >> -r--r--r-- 1 lewis users 14542313 Mar 11 2001 SqueakV3.sources >> $ cksum *sources >> 3417810574 14542313 SqueakV3.sources >> $ wc *sources >> 32 1898349 14542313 SqueakV3.sources >> >> In case you need the actual method source: >> includesKey: key >> "Answer whether the receiver has a key equal to the argument, key." >> >> self at: key ifAbsent: [^false]. >> ^true >> >> > >Thanks for the pointer....I checked my sources file and it's the same size >as yours! In fact, it works great with my 3.0 image...the problem must >lie within my 3.1 image.

1 0

[BUG][FIX] CommandShell bug in MVC
by David T. Lewis 29 Oct '01

29 Oct '01

CommandShellV1-3-1-MVC-patch Fixes a timing problem for CommandShell pipelines running in MVC. Apply this as an incremental change to CommandShell V1-3. Symptom: A command line such as '$ ls | edit' or '$ ls | inspect' did not work in an MVC shell window (Morphic was OK). The editor or inspector window would open and display an empty string. Cause: In MVC, the 'edit' and 'inspect' builtin commands scheduled an editor or inspector window to be opened at a later time using #addDeferredUIMessage:, and then cheerfully went on to close the input pipe for the command prior to the deferred action being executed. Fix: in a command pipeline, input to the 'edit' and 'inspect' shell builtin commands is now read completely prior to scheduling the editor or inspector window and closing the input pipe. This change set also fixes a harmless but unappealing buglet in which the last proxy in a command pipeline was sent the #value message twice.

1 0

RE: Lots of concurrency
by Joerg Beekmann 28 Oct '01

28 Oct '01

Yes it is a parallel world and I'll take your word for it that "people clearly don't think sequentially " but we also clearly structure our daily activities sequentially. Check anybodies to-do list. Some people, circus performers and manager, can juggle several activates. The fact they can is regarded as a notable feat, not the natural order of things. Moreover we model the world as system of interacting sequential agents. Other people out there are performing their own activities sequentially. So if we're going to use this analogy to design programming languages we want a language that provides for robust interaction of may sequential tasks operating in parallel, without the need for a lot of centralized control. The problem with today's state of the art seems to be that our software entities don't have sufficient understanding of the protocols required to operate in a parallel environment. In turn the environment doesn't have the requisite institutions and structures to allow independent parallel entities to operate. Joerg >>Whether people are "happier thinking sequentially" is a very important >>question. [As an aside, people clearly don't think sequentially - the brain >>has huge amounts of internal parallelism. And I think it is just an illusion >>that this parallelism is only at a low level (e.g. neurons). Read Minsky's >>Society Theory of Mind ( http://www.media.mit.edu/people/minsky/ ) for >>example.] TELUS Enterprise Solutions ....... Joerg Beekmann beekmann(a)columbusgroup.com t: 604 801 5758 ext. 300 c: 604 512 2557 Vancouver + Toronto + Calgary + Edmonton

8 11

Mental Threading
by Ed Heil 28 Oct '01

28 Oct '01

On Sun, Oct 28, 2001 at 12:03:44PM +0100, squeak-dev-request(a)lists.squeakfoundation.org wrote: > Message: 1 > Date: Sat, 27 Oct 2001 06:29:57 -0400 > To: squeak-dev(a)lists.squeakfoundation.org > From: Mark Guzdial <guzdial(a)cc.gatech.edu> > Subject: Re: Lots of concurrency > Reply-To: squeak-dev(a)lists.squeakfoundation.org > > "Sciences of the Artificial" is more on design than psychology. > "Protocol Analysis" with Ericsson is where I've heard more of the > evidence that the mind is single-threaded (e.g., similar to the > current work showing that driving and cell phone use reduces > attention to each), but I'd bet that "Models of Bounded Rationality" > has it, too. I think it's clear that the mind as a whole works in a massively parallel fashion, but that only a few threads at a time at most can be "conscious," "accessible to consciousness," or whatever term you want to use for that mental thingy that handles 7+/-2 chunks at once, maximum. Sources: vague memories of _The User Illusion_ by Tor Norretranders and _Wet Mind_ by Kosslyn & Koenig. -- ------------------------------------------------------ "A. We're not evil. B. We're not an empire." --Steve Ballmer of Microsoft, lying through his teeth ====================================================== Ed Heil ....................... uncorrected(a)yahoo.com ------------------------------------------------------

3 2

Re: Lots of concurrency
by Ken Kahn 28 Oct '01

28 Oct '01

Mark Guzdial wrote: > >And I think it is just an illusion > >that this parallelism is only at a low level (e.g. neurons). Read Minsky's > >Society Theory of Mind ( http://www.media.mit.edu/people/minsky/ ) for > >example. > > But also consider Herb Simon's arguments in opposition -- and Simon > has a lot more empirical evidence in his favor. I don't have an > opinion on which is right yet, but I don't think that this is a > settled point. > I'll agree it isn't settled. At the level of thoughts that I have self-awareness of, they seem pretty sequential but that is just the "tip of the iceberg". > > I really love Mitchel's MultiLogo work, but part of what I love about > it is his honesty in how *confusing* students found all the > concurrency. > Yes. I too find sequential languages with a few concurrency constructs confusing. I want languages designed from the bottom up to be concurrent. > > I don't find the "year of Logo programming" argument convincing. > There are too many studies (most prominently the Pea and Kurland > work, but even Idit Harel's and Yasmin Kafai's versions of ISDP) that > shows that not much gets learned in a year of programming. That deep > mindsets about the universe get changed in a single year is a > stretch. (For example, Idit's and Yasmin's studies have taken more > than a full year.) > Mitch writes "Several researchers (Pea et al. (1987), Bonar and Soloway (1985)) note that novice programmers (using traditional sequential languages) often assume parallelism where none exists." I have noticed this as well - especially in the context of object-oriented programming. Novices assume that each object is active and working independent of the others. Regarding whether a year is enough I think it rarely is. But as Mitch writes, "the students were part of a special computer-intensive environment, in which students worked at the computer for roughly an hour each day". And they probably were taught by MIT grad students. Most studies are of students taught by "ordinary" teachers in classes that meet less frequently. > The other examples (sports teams, traffic, etc.) seem more an > argument that students hold a centralized, sequential model of the > universe -- consider Mitchel's work with StarLogo and how hesitant > the students were to release the centralized models. > Yes, students hold centralized models (unfortunately). But I'm don't think they are sequential. They may all think that the bird in front is leading the flock but all the birds are flying at the same time. The centralized mindset interferes with seeing emergent phenomenon where there is no leader. > It should be noted that Mitchel's StarLogo work is a dissertation > about MENTAL MODELS OF CONTROL, *NOT* programming. I asked Mitchel > once about the interface that students used to StarLogo, and he told > me that he was it. None of his subjects actually wrote any of those > programs! Rather, they told Mitchel about their ideas, and he coded > them -- explaining what he was doing -- and then worked with the kids > to understand the results. It's important to note that the kids > didn't write the code. They might have been able to, but that hasn't > been tested As far as I know, there have been no empirical studies > of kids programming in StarLogo -- we don't know if it would work for > the average kid. So, we can't use StarLogo as an example of a > concurrent programming language that works for kids. > We probably can assume that a fair number of high school students can program StarLogo. A Google search 'starlogo "high school"' yields almost 400 hits; 'starlogo school' yields almost 1500. StarLogo once only ran on a million dollar Connection Machine so student access was probably limited. There are lots of wonderful StarLogo sample projects (and NetLogo too) but I don't like the SIMD flavor of the system. I would rather have the flexibility of having lots of autonomous communicating activities. Best, -ken kahn ( www.toontalk.com )

11 14

← Newer
1
...
7
8
9
10
11
12
13
...
60
Older →

Jump to page: