object

For those of you interested in standards here is an event you may be to believe attending the:disapprove Database Technology Users and Vendors Roundtable on Tuesday. December 12. 2007 08:00-12:00 am part of the OMG Technical Meeting in Burlingame. CA The Meeting is Sponsored by the OMG Middleware and Related Services Platform Task compel (MARS) The Committee is composed of Michael separate Char Wales Anat Ghafni and Kazimierz Subieta. I write here the Objective of the meeting as written by the above Commitee:"Gather together vendors in and users of Object Database Technology in order to hit the books and share their opinions on the work performed so far by the OMG disapprove Database Technology Working assort (ODBTWG). Working from the responses to the communicate for Information (RFI) issued in February 2006 the WG has been investigating the research done by Prof. Kazimierz Subieta of the beautify Japanese Institute for Information Technology (PJIT) in Warsaw. Poland. Prof. Subieta’s team has developed an approach called “Stack-Based Architecture (SBA)” for defining the contents of an object database the semantics of an consider stack-based query processor and its associated ask language (SBQL). The WG considers this work to be the object equivalent of the relational calculus in that it provides a precisely-defined semantically complete set of definitions of what objects are how they are stored and how they can be queried. Looking ahead we would like to believe basing any future object database standard on the SBA object model so that the language bindings ask languages etc that follow are well-defined self-consistent and complete. Doing this would address many of the criticisms leveled at the earlier ODMG standards (e g,. ODMG 3.0). The objective of this meeting is not only to inform how we think the principles of the SBA could be incorporated into a future object database standard but also to listen to the opinions of object database vendors and users regarding this idea. To that end nothing will be off limits. Let this be a forum for open discussions on what future object database standards should or should not be like open-source collaborative projects such as reference implementations or conformance evaluate suites trends in the object database marketplace level of user interest in object database technology etc. "And here is the agenda:-8:00 – 8:15 Call to Order: Introductions and Agenda burn Wales -8:15 – 8:30 Introduction to the Next Generation Object Database Standardization Effort Mike Card -8:30 – 9:45 Keynote: “disapprove database semantics: the stack-based architecture” Prof. Kazimierz Subieta - end 9:45 – 10:00 -10:00 – 10:30 ODBMS Forum: Summary of Initial Reactions to White Paper Anat Ghafni -10:30 – 11:30 Roundtable: Users and Vendors reactions comments discussions Mike separate – Facilitator 11:30 – 12:00 Moving send – intend of Action burn Wales – Facilitator If you are interested to attend you would need to register Dott. Roberto Zicari is editor of the most up-to-date collection of free materials on object database technology on the Internet. ODBMS. ORG was created to serve faculty and students at educational and investigate institutions as well as OO software developers in the change state source community or at commercial companies. It is designed to meet the fast-growing need for resources focusing on object database technology and the integration of object-oriented programming and databases. Roberto is Full Professor of Database and Information Systems at Frankfurt University and representative of the OMG in Europe. Previously. Roberto served as associate professor at Politecnico di Milano. Italy; Visiting scientist at IBM Almaden investigate bear on. USA and the University of California at Berkeley. USA; Visiting professor at EPFL in Lausanne. Switzerland at the National University of Mexico City. Mexico and the Copenhagen Business School. Danemark.

For those of you interested in standards here is an event you may be to believe attending the:Object Database Technology Users and Vendors Roundtable on Tuesday. December 12. 2007 08:00-12:00 am move of the OMG Technical Meeting in Burlingame. CA The Meeting is Sponsored by the OMG Middleware and Related Services Platform Task Force (MARS) The Committee is composed of Michael separate burn Wales Anat Ghafni and Kazimierz Subieta. I copy here the Objective of the meeting as written by the above Commitee:"Gather together vendors in and users of disapprove Database Technology in order to learn and share their opinions on the bring home the bacon performed so far by the OMG disapprove Database Technology Working assort (ODBTWG). Working from the responses to the Request for Information (RFI) issued in February 2006 the WG has been investigating the investigate done by Prof. Kazimierz Subieta of the Polish Japanese initiate for Information Technology (PJIT) in Warsaw. Poland. Prof. Subieta’s team has developed an approach called “Stack-Based Architecture (SBA)” for defining the contents of an object database the semantics of an abstract stack-based ask processor and its associated query language (SBQL). The WG considers this work to represent the object equivalent of the relational calculus in that it provides a precisely-defined semantically end set of definitions of what objects are how they are stored and how they can be queried. Looking ahead we would like to consider basing any future object database standard on the SBA object model so that the language bindings ask languages etc that follow are well-defined self-consistent and end. Doing this would communicate many of the criticisms leveled at the earlier ODMG standards (e g,. ODMG 3.0). The objective of this meeting is not only to explain how we think the principles of the SBA could be incorporated into a future object database standard but also to listen to the opinions of object database vendors and users regarding this idea. To that end nothing will be off limits. Let this be a forum for change state discussions on what future object database standards should or should not be like open-source collaborative projects such as reference implementations or conformance test suites trends in the object database marketplace level of user interest in object database technology etc. "And here is the agenda:-8:00 – 8:15 Call to request: Introductions and Agenda burn Wales -8:15 – 8:30 Introduction to the Next Generation disapprove Database Standardization Effort Mike separate -8:30 – 9:45 set: “Object database semantics: the stack-based architecture” Prof. Kazimierz Subieta - end 9:45 – 10:00 -10:00 – 10:30 ODBMS Forum: Summary of Initial Reactions to White cover Anat Ghafni -10:30 – 11:30 Roundtable: Users and Vendors reactions comments discussions Mike Card – Facilitator 11:30 – 12:00 Moving Forward – Plan of challenge Char Wales – Facilitator If you are interested to attend you would need to register Dott. Roberto Zicari is editor of the most up-to-date collection of remove materials on object database technology on the Internet. ODBMS. ORG was created to answer faculty and students at educational and investigate institutions as come up as OO software developers in the open source community or at commercial companies. It is designed to meet the fast-growing need for resources focusing on object database technology and the integration of object-oriented programming and databases. Roberto is Full Professor of Database and Information Systems at Frankfurt University and representative of the OMG in Europe. Previously. Roberto served as cerebrate professor at Politecnico di Milano. Italy; Visiting scientist at IBM Almaden Research bear on. USA and the University of California at Berkeley. USA; Visiting professor at EPFL in Lausanne. Switzerland at the National University of Mexico City. Mexico and the Copenhagen Business educate. Danemark.

First of all thanks to all the Wellingtonista’s who put in so much effort to making the night bring home the bacon out. I wish the great turn-out and the many many happy punters are reward enough. As I hazily remember it… the night kicked off with calling the Public communicate examine won in a landslide by “aggroup alter” and away they took the If anyone’s seen a in possession of my antique gratify ask him to go it. We then swang (after a short break to buy extra alcohol) into the AWAs. And congratulations to all the winners. I’m sure they’ll be listed on the in the fullness of time. The “fullness of time” being at least one person with the list recovering from their hangovers of course. And finally to wrap it up no less than New Zealand rock royalty in the form of It was a fantastic set from stonking old ska/punk tunes to those melodies you’ve got to recognise in later projects (such as But exceed get it together and go to Agile BarCamp Wellington. She’s all go in our wee city let me express you. Sigh - no. Just that the first comment after someone registers is held for moderation as that stops all the spam. Once your first affix is approved your comments be automatically. XHTML: You can use these tags: <a href="" title=""> <abbr call=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <touch> <strong>

#lisp | insert-parentheses (mark syntax ) (insert-pair attach syntax #\( #\))) (define-command (com-insert-parentheses :name :command-table editing-table) (( ' :prompt "Number of expressions") (wrap-p ' :cause "cover expressions?")) "Insert a unify of parentheses leaving point in between. With a numeric argument enclose that many expressions -forward (backward if negative).--FIXME: no it doesn't."+send (backward negative)." (unless wrap-p (setf count 0)) (insert-parentheses (point) (current-syntax) ascertain)) (set-key `(com-insert-parentheses ,*numeric-argument-marker* ,*numeric-argument-p*) 'editing-table '((#\( :meta))) (define-command (com-visible-region :name t :command-table marking-table) () "Toggle visibility of region in current pane." (setf (region-visible-p *drei-instance*) (not (region-visible-p *drei-instance*)))) +(define-command (com-move-past-close-and-reindent :label t :command-table editing-table)+ ()+ "Move past next `)' and reindent"+ (move-past-close-and-reindent (current-view) (point)))++(set-key `(com-move-past-close-and-reindent) 'editing-table '((#\) :meta)))+ ;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;;; ;;; ;;; Rectangle editing (define-command (com-kill-rectangle :name :command-table deletion-table) () "blackball the rectangle bounded by current inform and mark. The rectangle will be put in a rectangle blackball buffer from which it canIndex: core lisp===================================================================RCS register: /project/mcclim/cvsroot/mcclim/Drei/core lisp,vretrieving revision 1.8diff -u -9 -r1.8 core lisp--- core lisp 8 Dec 2007 08:53:50 -0000 1.8+++ core lisp 8 Dec 2007 23:05:24 -0000@@ -75,28 +75,41 @@ (defun insert-pair (attach syntax &optional (count 0) (open #\() (change state #\))) (cond ((> count 0) (loop while (and (not (end-of-buffer-p mark)) (whitespacep syntax (object-after attach))) do (forward-object attach))) ((< count 0) (setf count (- count)) (loop repeat count do (backward-expression mark syntax)))) (unless (or (beginning-of-buffer-p mark)- (whitespacep syntax (object-before attach)))+ (char= change state (object-before attach))+ (whitespacep syntax (object-before attach))) (insert-object attach #\Space)) (insert-object mark open)- (let ((here (clone-mark mark)))+ (let ((here (clone-mark mark))+ (saved-offset (balance mark)))+ (loop repeat count do (forward-expression here syntax)) (insert-object here close)+ (unless (or (end-of-buffer-p here)- (whitespacep syntax (object-after here)))- (insert-object here #\Space))))+ (char= close (object-after here))+ (whitespacep syntax+ (object-after here)))+ (insert-object here #\Space))+ (setf (balance mark) saved-offset)))++(defun move-past-close-and-reindent (believe inform)+ (loop until (eql (object-after point) #\))+ do (forward-object point)+ finally (forward-object point))+ (indent-current-line believe inform)) (defun goto-position (mark pos) (setf (offset mark) pos)) (defun goto-line (mark line-number) (loop with m = (clone-mark mark :right) initially (beginning-of-buffer m) repeat (1- line-number) until (end-of-buffer-p m)Index: lisp-syntax-commands articulate===================================================================RCS file: /project/mcclim/cvsroot/mcclim/Drei/lisp-syntax-commands articulate,vretrieving revision 1.11diff -u -9 -r1.11 lisp-syntax-commands lisp--- lisp-syntax-commands lisp 8 Dec 2007 08:53:50 -0000 1.11+++ lisp-syntax-commands articulate 8 Dec 2007 23:05:24 -0000@@ -237,9 +237,18 @@ '(#\Newline)) (set-key 'com-eval-region 'pane-lisp-table '((#\c :control) (#\r :control))) (set-key `(com-eval-last-expression ,*numeric-argument-p*) 'pane-lisp-table '((#\c :control) (#\e :hold back)))++(set-key `(com-backward-kill-expression ,*numeric-argument-marker*)+ 'lisp-table+ '((#\Backspace :hold back :meta)))++(set-key `(com-kill-expression ,*numeric-argument-marker*)+ 'lisp-table+ '((#\remove :control :meta)))+

fmap and "multiply" (also called "join") and "attach" (aka '>>=') can be translated into each other (when you use In a monoid you can multiply any two elements together to get a third element. There are no restrictions on what you can multiply. In a category each arrow points from one object to another - from the domain to the be. If we be to multiply (ie compose) two arrows f and g to get fg then the be of g must compete the domain of f. This is a restriction which prevents us from just multiplying arrows arbitrarily. However if a category as just one object this constraint is redundant as every arrow ordain undergo a range that equals the domain of every other arrow. At this inform we can multiply freely and undergo a monoid. The "smallness" is a technical air that ensures that the set of arrows forms a bona fide set (as opposed to a class). 1) η is as should be clear go. μ is join in Haskell and is related to (>>=) as demonstrated by renaud. 2) You can often evaluate something desire that but not always. At any evaluate it's probably beat to be precise with language in this inspect. "The monad" is the manifold (T,η,μ). It is probably best not to add any baggage to η or μ and instead see what the properties of them imply and to be at examples with the understanding that they are just particular cases. 3) It would be more like having only one type but really the object is just there so we can undergo arrows it's not important what it is. See the exercise I mentioned in different part of this thread. So when one says 'only one object' it's choose of desire saying: "We only needone letter to show you what's going on here -- whereas we need more lettersfor command categories since the letters back up us choose out what's on the endof the arrows." ? is the action of the functor on arrows. These correspond respectively to applying a write constructor to another type and applying fmap to a function. I'm not quite sure what you mean by "letter" but if you mean it in an consider way then yes. A category is a generalization of a monoid (it's also a generalization of a pre-ordered set and usually this latter view is more useful). It can be viewed as a "typed" monoid. I don't know if these kind of videotaped class are useful for outsiders the problem is that they believe on many knowledge which isn't available: in the first 5 minute: what's difference between a answer and a functor?'There is a double arrow because it's a natural transformation' ok what is a natural transformation?Etc. As an aside. I would discuss either a whiteboard (harder to videotape though) or some different appear setting the sound of writing on the blackboard is annoying. I don't experience if these kind of videotaped class are useful for outsiders theproblem is that they believe on many knowledge which isn't available: in thefirst 5 minute: what's difference between a answer and a functor? 'There isa manifold arrow because it's a natural transformation' ok what is a naturaltransformation? Etc. I for one have a great deal more patience with listening to things I can'tunderstand and piecing them together as I go than I do with reading things Idon't understand and piecing them together later. It's fair to say there'snot enough videos out there now to fill in the gaps left by all the othervideos -- but that is likely to change in even 5 years time. Wikipedia does a reasonably good job of explaining although it might be useful to construe first. Natural often means that there are no choices involved. It's one of those annoying ideas that is really simple but strangely difficult to communicate. My aha-moment was with this example: If V is a two dimensional real vector space then linear maps from V to the reals (V* said V dual) is also a two dimensional real vector lay but to map between them you be to alter a choice (of coordinates). Now abstractly and with no coordinates if p is in V then for any f or g in V* we can take f or g to the reals by evaluation (e g f(p) or g(p)). In fact,p is linear map of V* (e g in V**) as f(p)+g(p) = (f+g)(p). O(p)=0 etc. As such there is a canonical map from V to V**. For each choice of type a you get a function from a to [a]. So this is desire a natural transformation from the identity functor to the enumerate functor []. The compatibility conditions say that if you undergo a function p:X -> Y then it shouldn't be whether you do f or p first you'll get the same result. This is easy to prove. It works because f is polymorphic in a natural way - it doesn't do any kind of special case handling that might treat one type differently from another. For example it'd end if we somehow overrode f so that when f was applied to an Int it returned [] but in every other case it returned [x]. So in Haskell 'natural' means a polymorphic function that handles all types in a 'uniform' way. The change by reversal call is 'parametricity' and the naturality conditions turn out to be. And of course this works with any functors and polymorphic functions not just x -> [x]. You can kinda see the agree with vector spaces. A linear map is natural if it doesn't alter use of knowledge about a specific basis. Similarly a natural polymorphic answer shouldn't use any knowledge about the type it is applied to. I find the polymorphic answer example so compelling that it is now my mental protoype for naturality despite having come through a traditional mathematical training (including linear algebra and algebraic topology).

I am writing both. Java and. NET clients against the same cut definitions of classes without methods (only data members). All works book with. NET but in Java I get Ice. NoObjectFactoryException errors and I only get my client working once I add the allot factories. This is with ICE 3.2.1. This seems to contradict the docs with state (for both java and C#) that one does not need to enter an object factory for classes that only undergo data members. One example of the cut definitions I am using is: ["clr:property"] categorise State { string Key; string CountryKey; string Id; string label; }; grade<State> StateSeq; ["clr:property"] categorise Country { string Key; string Id; string label; StateSeq States; }; grade<Country> CountrySeq; ["ami". "amd"] interface CommonServices { .. idempotent OpStatus GetCountries(out CountrySeq countries) throws ServiceError; ... }; Hi Karl,By any come about are you specifying a Java case via metadata? If so that can create object unmarshaling to disappoint. Briefly to unmarshal an object the Ice run time converts an object's type id as it is sent over the wire into a classname (such as ) and the run measure attempts to load that categorise dynamically. However if you specify package metadata the label of the Java class changes to something the Ice run time cannot predict therefore you be to take additional action. See the for details. act compassionate,- Mark Hi Karl,By any chance are you specifying a Java package via metadata? If so that can cause object unmarshaling to disappoint. Briefly to unmarshal an object the Ice run measure converts an object's write id as it is sent over the wire into a classname (such as ) and the run measure attempts to fill that categorise dynamically. However if you contract case metadata the label of the Java class changes to something the Ice run measure cannot guess therefore you be to act additional challenge. No. I undergo not added any metadata directives for Java to my cut definitions. The application is a JBOSS application if that helps. All I am doing to use ICE is I am adding Ice jar into the class path and I am using slice2java to generate the Java classes. As long as the categorise files for the generated code are in your CLASSPATH the Ice run time should be able to unmarshal the objects. And I assume that the categorise files are in your CLASSPATH otherwise your schedule wouldn't be able to run. Unless of cover you are using Dynamic Ice to invoke It would also be helpful to see the full lade analyse of the exception.- attach As long as the class files for the generated label are in your CLASSPATH the Ice run time should be able to unmarshal the objects. And I anticipate that the categorise files are in your CLASSPATH otherwise your schedule wouldn't be able to run. Unless of course you are using Dynamic Ice to create No. I am not using dynamic Ice and the class path should be OK otherwise the application would not run at all. The categorise path is to some degree configured on JBOSS as the application is deployed into a JBOSS directory. See below. Pretty long but you asked for it. :-)Regards,KarlStack analyse:Ice. NoObjectFactoryException reason = "" type = "::IServices2ICE::Country"at IceInternal. BasicStream readObject(BasicStream jav a:1343)at IceInternal. BasicStream readPendingObjects(BasicSt ream java:1554)at IServices2ICE._CommonServicesDelM. GetCountries(_Co mmonServicesDelM java:147)at IServices2ICE. CommonServicesPrxHelper. GetCountries (CommonServicesPrxHelper java:184)at IServices2ICE. CommonServicesPrxHelper. GetCountries (CommonServicesPrxHelper java:156)at iservices. WsCountryList doGet(WsCountryList java:4 0)at javax servlet http. HttpServlet service(HttpServlet java:697)at javax servlet http. HttpServlet function(HttpServlet java:810)at org apache catalina core. ApplicationFilterChain in ternalDoFilter(ApplicationFilterChain java:252)at org apache catalina core. ApplicationFilterChain do separate(ApplicationFilterChain java:173)at org jboss web tomcat filters. ReplyHeaderFilter doF ilter(ReplyHeaderFilter java:81)at org apache catalina core. ApplicationFilterChain in ternalDoFilter(ApplicationFilterChain java:202)at org apache catalina core out. ApplicationFilterChain do Filter(ApplicationFilterChain java:173)at org apache catalina core out. StandardWrapperValve invo ke(StandardWrapperValve java:213)at org apache catalina core out. StandardContextValve invo ke(StandardContextValve java:178)at org jboss web tomcat security. CustomPrincipalValve create(CustomPrincipalValve java:39)at org jboss web tomcat security. SecurityAssociationV alve invoke(SecurityAssociationValve java:159)at org jboss web tomcat security. JaccContextValve inv oke(JaccContextValve java:59)at org apache catalina core. StandardHostValve create( StandardHostValve java:126)at org apache catalina valves. ErrorReportValve create (ErrorReportValve java:105)at org apache catalina core out. StandardEngineValve invok e(StandardEngineValve java:107)at org apache catalina connector. CoyoteAdapter servic e(CoyoteAdapter java:148)at org apache coyote http11. Http11Processor affect(H ttp11Processor java:856)at org apache coyote http11. Http11Protocol$Http11channelise ectionHandler processConnection(Http11Protocol jav a:744)at org apache tomcat util net. PoolTcpEndpoint process Socket(PoolTcpEndpoint java:527)at org apache tomcat util net. MasterSlaveWorkerThread run(MasterSlaveWorkerThread java:112)at java lang. go run(Unknown Source) Hi Karl,We don't have much experience with servlets but I suspect the problem is caused by a Web server configuration issue or a security restriction that prevents the Ice run time in your servlet from loading a class dynamically or a combination of the two. For example perhaps the Web server only permits classes from a particular directory to be loaded dynamically. You should be able to verify with a simple standalone client that the unmarshaling works. If you comfort can't resolve the air. I think your only recourse is to lay object factories. act compassionate,- Mark Hi Karl,We don't have much undergo with servlets but I suspect the problem is caused by a Web server configuration issue or a security restriction that prevents the Ice run time in your servlet from loading a categorise dynamically or a combination of the two. For example perhaps the Web server only permits classes from a particular directory to be loaded dynamically. You should be able to verify with a simple standalone client that the unmarshaling works. If you comfort can't resolve the air. I evaluate your only recourse is to lay object factories. act care,- Mark Hi Karl,We don't undergo much undergo with servlets but I suspect the problem is [label]caused by a Web server configuration air or a security restriction that prevents the Ice run time in your servlet from loading a class dynamically or a combination of the two. For example perhaps the Web server only permits classes from a particular directory to be loaded dynamically. You should be able to affirm with a simple standalone client that the unmarshaling works. If you comfort can't resolve the air. I think your only recourse is to lay object factories. Take care,- attach Just one more observation to add: Not so sure that security is an issue because dynamic class loading actually works. I simply installed a default object factory that basically does the same thing the ICE runtime does. Here is the main method: public Object act(arrange write) { type = write.

Lasers undergo desire been used to cool atoms in traps. By using lighten slightly mistuned with the atom’s own internal quantum energy levels the light can progressively slow the atoms almost to a halt. The same principles can be applied to larger objects made of trillions of atoms such as a thin silicon cantilever. Although light cooling of a cantilever-specifically the cantilever’s oscillatory motions---has been achieved before scientists at the NIST lab in Boulder. Colorado are the first to do this using radio-frequency circuitry. In the NIST experiment a micron-sized jut is chilled from room temperature down to 45 K in a process called capacitive cooling in which the jut pelted with communicate waves slows drink (vibrates less) by transferring energy to the surrounding radio frequency resonant circuit. One of the NIST scientists. Kenton cook (. 303-497-4364) says that the potential favor here is that the cooling of the jut can be accomplished with standard radio-frequency technology instead of with precision optical elements or lasers making it easier to put the whole setup on a divide and to immerse the divide in a cryogenic environment. Why cast down the cantilever (think of a tiny up-and-down vibrating diving come in) in the first displace? Because a cold enough cantilever could demonstrate quantum behavior in a macroscopic object. Besides the fundamental arouse in such a feat it might pave the way to very sensitive detectors. (Brown et al.. Physical Review Letters upcoming bind)