Search Results

Search found 1079 results on 44 pages for 'specification'.

Page 4/44 | < Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >

  • Anyone know why the Intel q9400 cpu is embedded?

    - by Wil
    I was just looking through the Intel site and I came across this - http://ark.intel.com/ProductCollection.aspx?familyID=28398 Anyone know why the 9400 has a tick in the embedded column? I have tried to contact Intel and not had a response. I have looked around but cannot find any additional reference and it seems to be available from shops just like any other CPU. Anyone have any ideas?

    Read the article

  • If you had two projects with the same specification and only one was developed using TDD how could you tell?

    - by Andrew
    I was asked this question in an interview and it has been bugging me ever since. You have two projects, both with the same specification but only one of these projects was developed using Test Driven Development. You are given the source for both but with the tests removed from the TDD project. How can you tell which was developed using TDD? All I was able to muster up was something about the classes being more 'broken up' in to smaller chunks and having more visible APIs, not my proudest moment. I would be very interested to hear a good answer to this question.

    Read the article

  • Should one use the legal "shall" in requirements documents and specification documents? [migrated]

    - by Billy ONeal
    At least in the US, "will" has replaced "shall" in most every context, with the notable exception of the "legal shall". Shall is used instead of will in legal documents to indicate a sense of obligation or requirement; e.g. "the defendant shall vacate the premises by October 16". In software, requirements documents and specification documents serve close to the same purpose as the aforementioned legal documents; does this mean shall should be used in a similar fashion as a result?

    Read the article

  • Le W3C valide une spécification sur la confidentialité des internautes proposée par Microsoft, la normalisation du standard commence

    Le W3C valide une nouvelle spécification proposée par Microsoft Pour lutter contre le traçage des internautes Un nouveau standard Web, fondé sur la fonctionnalité « Tracking protection » d'Internet Explorer 9 pour aider les utilisateurs à mieux contrôler le traçage de leur activité en ligne, avait été proposée par Microsoft au W3C. Elle vient d'être validée. Dans un billet, Microsoft affirme que l'organisme de normalisation responsable de la définition des standards du Web a récemment porté son attention sur la protection de la vie privée en ligne et a trouvé sa proposition « opportune et en phase avec ses propres objectifs et priorités ». Pour Dean Hachamovitch, vic...

    Read the article

  • WebGL 1.0, spécification finalisée pour la bibliothèques d'accélération 3D libre soutenue par Chrome, Firefox, Opera et Safari

    La spécification de WebGL 1.0 est finalisée Pour la bibliothèques d'accélération 3D libre soutenue par Chrome, Firefox, Opera et Safari Ces dernier temps, vous avez entendu parler de la mise en place de WebGL dans les nouvelles versions des différents navigateurs web. Comme vous devez déjà le savoir, Firefox 4, Google Chrome 9 (rendez-vous à l'adresse "about:flags" pour activer WebGL supportent la nouvelle technologie (également soutenue par Apple pour Safari et par Opera). Khronos, un consortium de constructeurs de cartes graphiques (NVidia, AMD, Imagination Technologies, ...) et de différents grand acteurs dans le monde de la 3D (Activision, Epic Games, ...) annonce dur...

    Read the article

  • « Responsive Images » : afficher des tailles d'images qui varient avec les écrans, le W3C publie un nouveau projet de spécification HTML

    HTML : afficher des tailles d'images différentes en fonction des écrans Le W3C publie un projet de spécification pour des « Responsive Images » L'heure est au « Responsive » pour le développement de sites Web. Après le Responsive Design ? qui fait varier le nombre de colonnes et modifie les contenus à afficher en fonction de la taille de la fenêtre ? voici le HTML Responsive Images Extension. Ce projet du W3C vise à normaliser la manière d'afficher des tailles d'images différentes en fonction de la taille de la fenêtre de navigation et de la résolution de l'écran. Concrètement, f...

    Read the article

  • System testing - making sure the system conforms to specification. Validation?

    - by user970696
    After weeks of research I have nearly completed my thesis, yet I am unable to clear up my confusion contained in all previous threads here (and in many books): During system testing, we check the system function against system analysis (functional system design) - but that would fit to a definition of verification according to many books. But I follow ISO12207, which considers all testing as validation (making sure work product meets requirement for intended use). How can I justify that unit testing or system testing is validation, even though when I check it against specification? Which fullfils the definiton of verification? When testing that e.g. "Save button" works, is it validation? This picture shows my understanding of V&V, so different from many other sources, including ISTQB etc. Essential problem I have is that a book using the same picture also states on another place that: test activities in the area of validation are usability, alpha and beta testing. For verification, testable system requirements are defined whose correct implementation can be tested through system tests. Isn't that the opposite of what the picture says? Most books present the following picture, where validation is just making sure that customer needs are satisfied. Mind you that according to ISO, validation activity is testing.

    Read the article

  • I am wondering how the Plural-Field generic is to be rendered in the REST OpenSocial 1.0 API specifi

    - by DaveGrahamOrg
    In the OpenSocial Data specificaiton 1.0 for a Person object (social profile data) it includes the use of a generic called Plural-Field. The spec can be found at: http://opensocial-resources.googlecode.com/svn/spec/1.0/Social-Data.xml#Person In the 1.0 data specification there is no XSD and no examples showing the use of this generic Plural-Field. After puzzeling the spec for some time I think I might understand the user of this generic. I was hoping that someone could confirm or correct my understanding. For example the accounts field is a generic Plural-Field<Account> while the activities field is Plural-Field <string>. Am I right in assuming that the result XML would look like: <accounts> <Plural-Field> <primary>true</primary> <type>ntlm</type> <value> <Account> <domain>MYDOMAIN</domain> <userid>MYDOMAIN\davegraham</userid> <username>davegraham</username> </Account> </value> </Plural-Field> <Plural-Field> <primary>false</primary> <type>claims</type> <value> <Account> <domain>i:0#.f|claimsDomain</domain> <userid>i:0#.f|claimsDomain|davegraham</userid> <username>davegraham</username> </Account> </value> </Plural-Field> </accounts> <activities> <Plural-Field> <primary>true</primary> <type>ntlm</type> <value>cycling</value> </Plural-Field> <Plural-Field> <primary>false</primary> <type>claims</type> <value>swiming</value> </Plural-Field> </activities> Am I right in my interpretation of the spec?

    Read the article

  • How to construct a flowchart/storyboard in a Func Spec

    - by PeterQ
    Hey I'm a bit embarrassed to write a post on this topic, but I would appreciate the help. At my school, the CS kids (myself included) have created a nice, little program that is built for incoming Chem/Bio students. It consists of several modules that reviews topics they should have a firm grasp on before they start their classes. It's a nice tool since it cuts down on reviewing the material in class but also allows the students to do a quick diagnostic to fix any problems. Now, I'm in charge of constructing a simple interface that reports on the progress of the group and individual students. The interface is pretty simple. It's just a window that pops up, and it has three tabs: the first tab is a "cumulative" report of all of students. The secnod tab has a drop down box that lists the students and once a student is selected, a report for him/her comes up. And the third tab is simply a description of all of the terms used in the 1st and 2nd tabs. Now, I'm trying to be a good CS student and write a func. spec for my interface. My problem comes with the fact that I'd like to insert a little flowchart using Visio. Problem is, and I'm quite embarrassed to admit this, I don't know how to construct the flowchart/storyboard. For instance, I know I start with a "Start/Click Icon" in a rectangle. Then where do I go? Do I draw three arrrows (one going to each tab) and then describing what goes on? In tab one, the only thing that happens is that the user will select a "sort" method in the drop down box. This will sort the list. The End. Similarly, if the user selects the second tab, then he will go to a drop down box with the student names. Selecting a name will bring up student info. And the third tab is just a list of unfamiliar terms coming from the first or second tab. I wanted to storyboard/flowchart what I'm doing, but I'm unclear how to go about it. Any help would be appreciated! (Just to clarify, I'm not having trouble with using Visio, but I don't know how one goes about construct a storyboard or determining the procedure for constructing one)

    Read the article

  • MS DOS function like ipconfig to get system performance specs?

    - by JustADude
    I am aware of MSINFO32, but I'm wondering if there is a MS DOS command similar to ipconfig in order to get system specifications? I would like for the system specifications to be displayed in the MS DOS prompt. I would like to see at least: CPU RAM BUS speed Thanks for any insights. Edit: I am unable to install any other software, so just have to use existing DOS programming commands to extract this information. Thank you again. 2nd Edit: Whoops. Using Windows XP and Windows Vista.

    Read the article

  • What's the best way to link formal specs to JIRA enchancement requests?

    - by Adam
    What's the best way to link formal specs to JIRA enhancement requests? I want to track changes to specifications using JIRA. Ideally, I'd like to refer to a functional ID reference in a JIRA ticket (e.g. MYAPPAPPROVAL LOGICMAIN SCREEN), so that program managers can retrospectively categorise defects. The reason for this, is so that QA scripts and documentation tickets can be searched/categorised meaningfully in the tracking system. There seems to be a million possible ways to do this, e.g. should I write a custom component to select functional IDs from a tree? should I write the specs in confluence, or another CMS with a TrackBack facility? should I include a link to the documentation URL? should I use some other 3rd party plugin application? should I use some Atlassian application that i'm unaware of? am I using the wrong tracking tool/process to measure spec growth? What's the best way, in your experience?

    Read the article

  • « Le rejet des DRM risque de cloisonner le Web » pour le PDG du W3C, qui trouve que la spécification EME est un juste compromis

    Le W3C étudie une norme pour la lecture du contenu protégé dans le HTML5 qualifiée de « contraire à l'éthique » par un membre du consortiumDes développeurs de Google, Microsoft et Netflix ont proposé une nouvelle norme pour le HTML5.Le futur standard du Web HTML5 qui est de plus en plus utilisé au détriment des technologies comme Flash ou Silverlight souffre encore de quelques manquements par rapport à celles-ci. C'est le cas par exemple pour la lecture du contenu vidéo protégé.Une nouvelle proposition a été faite au W3C par David Dorwin (Google), Adrian Bateman (Microsoft) et Mark Watson (Netflix) pour permettre au HTML5 de lire du contenu protégé DRM (Digital rights management ).Bapti...

    Read the article

  • What do you think of the following job specification?

    - by m.edmondson
    Just received this out of the blue from a recruiter - a number of things stand out to me: PERSON PROFILE Hard working - with a stay until the job in done mentality Thrive on the pressure of tight weekly development deadlines Good attention to detail to ensure bug free development Ability to test all development work from user's perpective Ability to think like a user as well as a developer Good communication skills to understand new funcationality and bugs Flexibility to contribute outside main responsbilities when needed. BENEFITS Salary dependant on skills Contributary Pension with 4% contribution from employer (after 1 year of service) Private Healthcase (after 1 year of service) 20 days holiday + 3-4 days holiday between Christmas and New year - 1 day extra holiday available each quarter you don't have a day off sick (and an additional day if you are not off sick for the whole year ). Would you want to work here? From what I can see they want a work-a-holic who will crawl out of his death bed in order to not lose holiday entitlement.

    Read the article

  • How do I convince my team that a requirements specification is unnecessary if we adopt user-stories?

    - by Nupul
    We are planning to adopt user-stories to capture stakeholder 'intent' in a lightweight fashion rather than a heavy SRS (software requirements specifications). However, it seems that though they understand the value of stories, there is still a desire to 'convert' the stories into an SRS-like language with all the attributes, priorities, input, outputs, source, destination etc. User-stories 'eliminate' the need for a formal SRS like artifact to begin with so what's the point in having an SRS? How should I convince my team (who are all very qualified CS folks by the way - both by education and practice) that the SRS would be 'eliminated' if we adopted user-stories for capturing the functional requirements of the system? (NFRs etc can be captured too, but that's not the intent of the question). So here's my 'work-flow' argument: Capture initial requirements as user-stories and later elaborate them to use-cases (which are required to be documented at a low level i.e. describing interactions with the UI prototypes/mockups and are a deliverable post deployment). Thus going from user-stories to use-cases rather than user-stories to SRS to use-cases. How are you all currently capturing user-stories at your workplace (if at all) and how do you suggest I 'make a case' for absence of SRS in presence of user-stories?

    Read the article

  • How does Trash Can works? Where can i find official specification / documentation / reference about it?

    - by MestreLion
    When trying to manage trash can from mounted NTFS volumes, I ended up reading FreeDesktop.org's reference on it. Poking around and doing some tests, I realized Ubuntu/Gnome does not follow the specs 100%. Here's why: For non-/ partitions, it always use <driveroot>/.Trash-<uid>, It never used <driveroot>/.Trash/<uid>, even when i created it in advance. While this works, its annoying: if i have 15 users, i end up with 15 /.Trash-xxx folders in my drive, while the other approach would still give a single folder (with 15 sub-folders). That "pollution" in my drives is very unpleasant. And specs say "If an $topdir/.Trash directory is absent, an $topdir/.Trash-$uid directory is to be used". Well, it IS present, so why it never uses it? root trash does not work, at least not out of the box. Open nautilus as root and click on trash, it gives error. Try to delete any file, it says "it cant move to trash". Ok, i know this can be fixed by creating /root/.local/share. But specs says "A “home trash” directory SHOULD be automatically created for any new user. If this directory is needed for a trashing operation but does not exist, the implementation SHOULD automatically create it, without any warnings or delays.". Why error then? Bug? Why do i must change /etc/fstab entries for mounted volumes, adding options like uid and guid, if the volumes are already mounted as RW for everyone? These are just some examples of deviation from standard. So, the question is: "If Ubuntu does not adhere 100% to the spec, HOW exactly does the trash work? WHERE can i find technical reference about Ubuntu's implementation of the trash?" By the way: if Ubuntu does happen to follow specs, please tell me what am i doing wrong, specially regarding the /.Trash-<uid> vs /.Trash/<uid> issue. Thanks! EDIT: Some more info: If a given fs has no support for sticky bit (VFAT, NTFS), it probably dont have for permitions either (at least VFAT surely doesnt). So what prevents one user for purging / restoring other users ./Trash-xxx ? If one can read/write his own Trash, he can also do the same for the whole drive, including other's trashes, isnt it? Or does Gnome has any "extra" protection on ./Trash-xxx folders on VFAT/NTFS fs? If Linux can "emulate" file permitions on NTFS mounting by editing /fstab uid and gid options, can it also "emulate" the sticky bit? I would really want to use /.Trash/xxx format... For the root issue: for the / partition, i can trash as root, and it goes to /root/.local/shate/Trash. But if i click on Nautilus "Trash" (as root), i get an error. Dont you? So files are correctly trashed, but i cant access it. All i can do is manually "purge" them (by deleting files on /root/.local/shate/Trash), but restoring would be very tricky (opening info files and manually moving, etc) For non-/ partitions (or at least for VFAT/NTFS), I can not even trash as root: it does not create a ./Trash-0 folder, it simply says "Cannot trash, want to permantly delete?" Why? About fstab: i use it for a permanent mount for my NTFS partitions. I have several, and if not "pre-mounted" they really cluttter desktop and/or Nautilus. Id rather have it pre mounted, integrated in my fs, in mounts like /data , /windows/xp , /windows/vista , and so on, and leave /media and its "mount/unmount" flexibility just for truly removable drives Si, if Ubuntu/Gnome truly follow the spec, is there any way to fix the root issues and to "emulate" the sticky bit for (at least) my fstab'ed NTFS fixed partitions?

    Read the article

  • How does the Trash Can work, and where can I find official documentation, reference, or specification for it?

    - by MestreLion
    When trying to manage trash can from mounted NTFS volumes, I ended up reading FreeDesktop.org's reference on it. Poking around and doing some tests, I realized Ubuntu/Gnome does not follow the specs 100%. Here's why: For non-/ partitions, it always uses <driveroot>/.Trash-<uid>, It never used <driveroot>/.Trash/<uid>, even when i created it in advance. While this works, it's annoying: if I have 15 users, I end up with 15 /.Trash-xxx folders in my drive, while the other approach would still give a single folder (with 15 sub-folders). That "pollution" in my drives is very unpleasant. And specs say "If an $topdir/.Trash directory is absent, an $topdir/.Trash-$uid directory is to be used". Well, it IS present, so why does it never use it? root trash does not work, at least not out of the box. Open nautilus as root and click on trash; it gives an error. Try to delete any file, it says "it can't move to trash". Ok, I know this can be fixed by creating /root/.local/share. But specs says "A “home trash” directory SHOULD be automatically created for any new user. If this directory is needed for a trashing operation but does not exist, the implementation SHOULD automatically create it, without any warnings or delays.". Why the error then? Bug? Why must I change /etc/fstab entries for mounted volumes, adding options like uid and guid, if the volumes are already mounted as RW for everyone? These are just some examples of deviation from the standard. So, the question is: "If Ubuntu does not adhere 100% to the spec, HOW exactly does the trash work? WHERE can i find a technical reference for Ubuntu's implementation of the trash?" By the way: if Ubuntu does happen to follow specs, please tell me what I am doing wrong, especially regarding the /.Trash-<uid> vs /.Trash/<uid> issue. Thanks! EDIT: Some more info: If a given fs has no support for the sticky bit (VFAT, NTFS), it probably doesn't have for permissions either (at least VFAT surely doesn't). So what prevents one user from purging / restoring other users' ./Trash-xxx ? If one can read/write his own Trash, one can do the same for the whole drive, including other's trashes, correct? Or does Gnome have some kind of "extra" protection on ./Trash-xxx folders on VFAT/NTFS fs? If Linux can "emulate" file permissions on NTFS mounting by editing /fstab uid and gid options, can it also "emulate" the sticky bit? I would really prefer to use /.Trash/xxx format... For the root issue: for the / partition, I can use trash as root, and it goes to /root/.local/shate/Trash. But if I click on Nautilus "Trash" (as root), I get an error. Don't you? So files are correctly trashed, but I can't access it. All I can do is manually "purge" them (by deleting files on /root/.local/shate/Trash), but restoring would be very tricky (opening info files and manually moving, etc.). For non-/ partitions (or at least for VFAT/NTFS), I can not even use trash as root: it does not create a ./Trash-0 folder, it simply says "Cannot trash, want to permanently delete?" Why? About fstab: i use it for a permanent mount for my NTFS partitions. I have several, and if not "pre-mounted" they really clutter the desktop and/or Nautilus. I'd rather have it pre-mounted, integrated in my fs, in mounts like /data , /windows/xp , /windows/vista , and so on, and leave /media and its "mount/unmount" flexibility just for truly removable drives. So, if Ubuntu/Gnome truly follows the spec, is there any way to fix the root issues and to "emulate" the sticky bit for (at least) my fstab'ed NTFS fixed partitions?

    Read the article

  • New iPad vs. iPad 2–Side by side comparison of hardware specification [Infographic]

    - by Gopinath
    Apple released the 3rd generation of iPad on March 7th with spectacular hardware and software specs. The new iPad is the most advanced tablet available in the market with not much of competition. The closest competitor to the new iPad is not from Android or RIM or Amazon as they are no where close to the standards of the new iPad . But the competitor is none other than previous generation of iPad 2. In order to help you decide which Apple tablet suits your requirements here is an infographic comparing the iPad  with iPad 2

    Read the article

  • What is the best HTML specification to be used as of Q1 2011?

    - by Rob McKinnon
    While developing a web application, what is the best spec to use? HTML4.01 HTML5 XHTML trans XHTML1.1 I was taught to use XHTML1.0 strict in uni and to avoid applet/iframe/tables(except in forms). I noticed that some deprecated tags are available in HTML5. Is it safe to code in HTML5? If so should I use target='', and the aforementioned tags? I have noticed that there are many alternatives to choose from including canvas, object. I have no preference, although Iframe tags are being dispensed from sources like Facebook/Google/etc. What would be the best avenue to take for Spec as of now(Feb 2011)?

    Read the article

  • Eclipse Kepler apporte le support de Java EE 7, sortie simultanée entre la spécification et l'environnement de développement de la fondation Eclipse

    Une nouvelle version d'Eclipse est disponible. Elle porte le nom de Kepler. Cette version marque la fin officielle du support de la branche 3.x d'Eclipse par la Fondation. Elle continue donc sur la lancée de Juno.Des informations supplémentaires sur les nouveautés de cette version sont disponibles à cette adresse : notes pour la version 4.3.Le projet Kepler se compose de 72 projets (114 en comptant les sous-projets), pour un total d'environ 58 millions de lignes de code par 428 committers. 5 projets ont rejoint le « simulatenous release train » : EMF Diff/merge, Sphinx, Stardust, Hudson et Maven integration pour WTP (Web Tools P...

    Read the article

  • C++14 : le draft final a été publié, découvrez son contenu et ce qu'apporte la nouvelle spécification

    C++14 : le draft final a été publié Découvrez son contenu Le draft final de la nouvelle version du C++14 est publié. La seconde bonne nouvelle, c'est que les compilateurs les plus utilisés supportent déjà cette nouvelle version. Voici ce qu'elle nous apporte : N3323 - Correction de certaines conversions contextuelles du C++ : améliorations du comportement de conversion à un unique opérateur, alors que la conversion d'une valeur de la classe vers le type spécifié par le contexte est possible...

    Read the article

  • The long road to bug-free software

    - by Tony Davis
    The past decade has seen a burgeoning interest in functional programming languages such as Haskell or, in the Microsoft world, F#. Though still on the periphery of mainstream programming, functional programming concepts are gradually seeping into the imperative C# language (for example, Lambda expressions have their root in functional programming). One of the more interesting concepts from functional programming languages is the use of formal methods, the lofty ideal behind which is bug-free software. The idea is that we write a specification that describes exactly how our function (say) should behave. We then prove that our function conforms to it, and in doing so have proved beyond any doubt that it is free from bugs. All programmers already use one form of specification, specifically their programming language's type system. If a value has a specific type then, in a type-safe language, the compiler guarantees that value cannot be an instance of a different type. Many extensions to existing type systems, such as generics in Java and .NET, extend the range of programs that can be type-checked. Unfortunately, type systems can only prevent some bugs. To take a classic problem of retrieving an index value from an array, since the type system doesn't specify the length of the array, the compiler has no way of knowing that a request for the "value of index 4" from an array of only two elements is "unsafe". We restore safety via exception handling, but the ideal type system will prevent us from doing anything that is unsafe in the first place and this is where we start to borrow ideas from a language such as Haskell, with its concept of "dependent types". If the type of an array includes its length, we can ensure that any index accesses into the array are valid. The problem is that we now need to carry around the length of arrays and the values of indices throughout our code so that it can be type-checked. In general, writing the specification to prove a positive property, even for a problem very amenable to specification, such as a simple sorting algorithm, turns out to be very hard and the specification will be different for every program. Extend this to writing a specification for, say, Microsoft Word and we can see that the specification would end up being no simpler, and therefore no less buggy, than the implementation. Fortunately, it is easier to write a specification that proves that a program doesn't have certain, specific and undesirable properties, such as infinite loops or accesses to the wrong bit of memory. If we can write the specifications to prove that a program is immune to such problems, we could reuse them in many places. The problem is the lack of specification "provers" that can do this without a lot of manual intervention (i.e. hints from the programmer). All this might feel a very long way off, but computing power and our understanding of the theory of "provers" advances quickly, and Microsoft is doing some of it already. Via their Terminator research project they have started to prove that their device drivers will always terminate, and in so doing have suddenly eliminated a vast range of possible bugs. This is a huge step forward from saying, "we've tested it lots and it seems fine". What do you think? What might be good targets for specification and verification? SQL could be one: the cost of a bug in SQL Server is quite high given how many important systems rely on it, so there's a good incentive to eliminate bugs, even at high initial cost. [Many thanks to Mike Williamson for guidance and useful conversations during the writing of this piece] Cheers, Tony.

    Read the article

  • The long road to bug-free software

    - by Tony Davis
    The past decade has seen a burgeoning interest in functional programming languages such as Haskell or, in the Microsoft world, F#. Though still on the periphery of mainstream programming, functional programming concepts are gradually seeping into the imperative C# language (for example, Lambda expressions have their root in functional programming). One of the more interesting concepts from functional programming languages is the use of formal methods, the lofty ideal behind which is bug-free software. The idea is that we write a specification that describes exactly how our function (say) should behave. We then prove that our function conforms to it, and in doing so have proved beyond any doubt that it is free from bugs. All programmers already use one form of specification, specifically their programming language's type system. If a value has a specific type then, in a type-safe language, the compiler guarantees that value cannot be an instance of a different type. Many extensions to existing type systems, such as generics in Java and .NET, extend the range of programs that can be type-checked. Unfortunately, type systems can only prevent some bugs. To take a classic problem of retrieving an index value from an array, since the type system doesn't specify the length of the array, the compiler has no way of knowing that a request for the "value of index 4" from an array of only two elements is "unsafe". We restore safety via exception handling, but the ideal type system will prevent us from doing anything that is unsafe in the first place and this is where we start to borrow ideas from a language such as Haskell, with its concept of "dependent types". If the type of an array includes its length, we can ensure that any index accesses into the array are valid. The problem is that we now need to carry around the length of arrays and the values of indices throughout our code so that it can be type-checked. In general, writing the specification to prove a positive property, even for a problem very amenable to specification, such as a simple sorting algorithm, turns out to be very hard and the specification will be different for every program. Extend this to writing a specification for, say, Microsoft Word and we can see that the specification would end up being no simpler, and therefore no less buggy, than the implementation. Fortunately, it is easier to write a specification that proves that a program doesn't have certain, specific and undesirable properties, such as infinite loops or accesses to the wrong bit of memory. If we can write the specifications to prove that a program is immune to such problems, we could reuse them in many places. The problem is the lack of specification "provers" that can do this without a lot of manual intervention (i.e. hints from the programmer). All this might feel a very long way off, but computing power and our understanding of the theory of "provers" advances quickly, and Microsoft is doing some of it already. Via their Terminator research project they have started to prove that their device drivers will always terminate, and in so doing have suddenly eliminated a vast range of possible bugs. This is a huge step forward from saying, "we've tested it lots and it seems fine". What do you think? What might be good targets for specification and verification? SQL could be one: the cost of a bug in SQL Server is quite high given how many important systems rely on it, so there's a good incentive to eliminate bugs, even at high initial cost. [Many thanks to Mike Williamson for guidance and useful conversations during the writing of this piece] Cheers, Tony.

    Read the article

  • Is Mac OS X a licensed Unix or Unix-like clone that conforms to Unix specification?

    - by KMC
    Is Mac OS X developed on a licensed Unix or is it a Unix-like clone that, unlike Linux, conforms to Unix specification well enough to be registered as a Unix OS. Not until Leopard, Mac OS X did not gain the Unix certification. But in Leopard, Terminal still print: GNU bash, version 3.2.48(1)-release (x86_64-apple-darwin10.0) But GNU is GNU's not Unix, and Mac OS X is registered as Unix. That gets me confused whether OS X is unix or unix-like. In other words, is OS X written on top of Unix or a re-write of Unix that is as Unix as it can possible be. May be along the answer someone can provide lineage or other background information. I would also recommend reading How Unix is Mac OS X.

    Read the article

< Previous Page | 1 2 3 4 5 6 7 8 9 10 11 12  | Next Page >