Slashdot Mirror


Something For (Almost) Every Developer

First up, reader martinjlogan sends along a tutorial for setting up a workable Erlang/OTP development environment on a Mac. Next, reader acid06 notes news of Perl 5.12, including what may be the first delivered fix for the Y2K38 bug. (Hit the Read More link below for some details on Perl's new release strategy.) "After two years of development, the new major version of Perl is now available. Notable new features are: better Unicode support, proper support for time after the Y2038 barrier, new APIs to allow developers to extend Perl with 'pluggable' keywords and syntax, warnings for deprecated features and more. From the linked post: You can get it from the CPAN right now or wait for a platform-specific release (such as Strawberry Perl for Windows)." Finally, from reader snydeq: "InfoWorld's Martin Heller provides an in-depth review of Visual Studio 2010 and finds Microsoft taking several large steps away from its legacy IDE code. 'Visual Studio 2010 is a major upgrade in functionality and capability from its predecessor. Developers, architects, and testers will all find areas where the new version makes their jobs easier. Despite the higher pricing for this version, most serious Microsoft-oriented shops will upgrade to Visual Studio 2010 and never look back,' Heller writes. Chief among the improvements are Microsoft's revamping the core editing and designer views to use WPF, its overhaul of IntelliSense and support for test-driven development, and its intelligent support for multiple versions of the .Net Framework."
Re: Perl. This release cycle marks a change to a time-based release process. Beginning with version 5.11.0, we make a new development release of Perl available on the 20th of each month. Each spring, we will release a new stable version of Perl. One month later, we will make a minor update to deal with any issues discovered after the initial ".0" release. Future releases in the stable series will follow quarterly. In contrast to releases of Perl, maintenance releases will contain fixes for issues discovered after the .0 release, but will not include new features or behavior.

263 comments

  1. VS upgrade cycle by QuantumG · · Score: 4, Insightful

    most serious Microsoft-oriented shops will upgrade to Visual Studio 2010 and never look back

    Of course, implying that you're not a serious Microsoft-oriented shop if you don't upgrade. This is the exact opposite of the case. As Microsoft regularly changes stuff in VS that no-one wants, most people don't upgrade until necessity forces it on them. It's entirely network effects. If you're using precompiled third party libraries and they upgrade, chances are you'll be forced to upgrade. If Microsoft made it easier to use the new IDE without upgrading the compilers, the standard lib, the header install, etc, I imagine more people would accept the feature improvements (and the bug fixes!) to the IDE without trepidation.

    --
    How we know is more important than what we know.
    1. Re:VS upgrade cycle by Anonymous Coward · · Score: 1, Insightful

      He was implying no such thing.

      Most serious Microsoft-oriented shops will do X =/=> All shops who do not do X are not both serious and Microsoft-oriented.

      most people don't upgrade until necessity forces it on them

      Not my experience when it comes to Visual Studio.

      If Microsoft made it easier to use the new IDE without upgrading the compilers, the standard lib, the header install, etc, I imagine more people would accept the feature improvements (and the bug fixes!) to the IDE without trepidation.

      Bad, bad, bad, bad, bad idea.

    2. Re:VS upgrade cycle by AndrewStephens · · Score: 4, Informative

      Agreed. I work in a "serious Microsoft shop" and we have just migrated our projects to VS2008. Experience has taught us that although the Microsoft Dev environments are of high quality, for the first 12 months there will be service packs and patches. We do not want to have to migrate our whole team and our projects every 3 months just to keep up.

      That said, I am looking forward to using VS2010 eventually. I couldn't care less about .NET but the new C++ language features are neat.

      --
      sheep.horse - does not contain information on sheep or horses.
    3. Re:VS upgrade cycle by QuantumG · · Score: 5, Insightful

      Bad, bad, bad, bad, bad idea.

      I think its a great idea.. there's no reason why the IDE release cycle has to be tied to the compiler release cycle.. except that Microsoft likes the lockin.

      --
      How we know is more important than what we know.
    4. Re:VS upgrade cycle by jellomizer · · Score: 1

      The problem you are facing is the fact that it isn't the serious Microsoft-oriented shop don't want to upgrade but they're programming skills are at a level where they can't upgrade. Using 3rd party libraries are always risky, and should be avoided unless it is really doing something that will take you much longer to accomplish. However a lot of the Microsoft shops have very poor programers or the odd hack who thinks he is a programmer by glueing a bunch of 3rd party tools even ones that already come with the default library. 3 or 4 grid controls, a Menu Bar Control or two... 3 or 4 image controls... So they end up with a program that they cannot upgrade. But this isn't really the fault of Microsoft though... It is just because of the bad programmers. If you see a VS.NET application written by a pro it is a clean fast and easily upgradable system.

      I Program in many different languages and I keep Visual Studio Ready for the jobs it does well. I remember startling a bunch of Java only programmers on how fast I was able to get my system to talk to their system and I was able to support the new features that they had on their spec list before they actually finished programming it themselves. That said when I need to process text data I prefer Python as it gets the job done better for that job.

      If you are willing to stop getting all huffy puffy, emotional and political about you're programming languages you will find there are a lot of better ways to get a lot of your work done.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    5. Re:VS upgrade cycle by h4rr4r · · Score: 1

      Using 3rd party libraries are always risky, and should be avoided unless it is really doing something that will take you much longer to accomplish.

      So reinventing the wheel is a good thing?
      I do not do MS related development, but from where I stand reinventing the wheel is nuts. Do you think all of CPAN is there just for the heck of it?

    6. Re:VS upgrade cycle by Sponge+Bath · · Score: 2, Insightful

      ...we have just migrated our projects to VS2008.

      They can have my MSVC 6 when they pry it from my cold dead GLURK, gaaasssspp...

      One problem I've had is the new redistributable requirements of each release. Not insurmountable, but a real concern when considering an upgrade.

    7. Re:VS upgrade cycle by yuriks · · Score: 5, Informative

      Actually, one of the features of 2010 is that you can target old compiler versions (starting with VS2008) with the new IDE.

    8. Re:VS upgrade cycle by QuantumG · · Score: 1

      reinventing the wheel

      is a kinda naive way to refer to horizontal integration of the industry.. and considering that you were replying to a guy who I thought was very naive, I'm overdosing on naivety here. Vertical integration has some great virtues but is not representative of a mature industry.

      --
      How we know is more important than what we know.
    9. Re:VS upgrade cycle by DAldredge · · Score: 1

      What does regularly change in VS that no one wants?

    10. Re:VS upgrade cycle by The+Snowman · · Score: 1

      Using 3rd party libraries are always risky, and should be avoided unless it is really doing something that will take you much longer to accomplish.

      ...unless you want to finish your project in half the time.

      Third party libraries, frameworks, and toolkits exist for a purpose: to make your life easier.

      Do you program GUIs by manipulating pixels in the video frame buffer? Do you manage files by communicating directly with the file system layer of the kernel? Interpret mouse clicks by talking to the mouse driver?

      I bet you use third party libraries all the time without even knowing it.

      --
      24 beers in a case, 24 hours in a day. Coincidence? I think not!
    11. Re:VS upgrade cycle by jellomizer · · Score: 2, Interesting

      I think CPAN is there to make you go nuts trying to get a downloaded Perl Script to work as you dig around trying to find the right libraries to make the freakin program work... But I am not a fan of Perl so I digress...

      The 3rd party problem isn't a case of reinventing the wheel. But getting new wheels that already exist just because they are slightly different, eg a grid control that automaticly populates from the database vs. Loading it from 3 other commands...

      But... Sometimes for programming reinventing the wheel is a good thing. Why...

      1. Maintenance. A 3rd party control means code you probably wont touch... Even if it is open source you will not probably spend to much time reworking it unless you really have to
      2. Portability. Just as the Grand Parent mentioned moving to the next version of you developer system may break the third party tool, while you own version can be upgraded much easier, also if you were say moving from a windows form based app to a web app you already have the logic there to port you app over.
      3. Knowlege. Black Boxes in your knowledge and your product can be a bad thing. If you do it yourself you know how it works, if you don't and you find a problem you look and feel stupid
      4. Your License. If you get a GNU 3rd party license you have to follow the GNU. If you get a closed source license you need to follow their rules... You make it yourself you can license it any way you choose, as well as the rest of your app.
      5. Optimal Solution. Your code will do just as you need it to do. No more, No Less 3rd party tools either have you taking out a feature you would like. Or just do a lot more then you need causing issue later on.
      6. Easier for others to follow. If someone else is tracing you code and you get a 3rd party tool, it is either a black box application that you are referencing where you need to break your trace to see what the heck it is for. Or there is source and you go in there and it being code by a different person it is like you are in Wonderland were all coding styles have changed.
      7. Deployment. If possible you can get that app to run as a single unit deploying it is much easier as say for windows no registering a bunch of crap.
      8. Cleaning. If you have 3rd party tools that have been there for a while, you really don't want to get rid of them without knowing how it will effect your other program. Your own code it is a bit easier
      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    12. Re:VS upgrade cycle by jellomizer · · Score: 1

      They are not 3rd party tool... They are the developments language mainline tools...
      There is a difference.
      Mainline tools you know they are there and if you move to a different development environment they are still there. 3rd party tool you take your code and move it to a new system it will not run until you reinstall the other tools.

      And as I stated in my post that you added as the quote "unless it is really doing something that will take you much longer to accomplish."

      Meaning that 3rd party tools are not evil however need to be chosen and used with care once you found out that there is no other easy option with the mainline tools.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    13. Re:VS upgrade cycle by LostCluster · · Score: 1

      This is the fourth release of Visual Studio (.NET, 2005, 2008, 2010) since VB6 and while many shops clung on to VB6 rather than convert to .NET, this is the point where a code rewrite is starting to make sense and it's time to get with the latest. 2010 has been available in a time-limited public beta and a non-limited MSDN release for months now... this year seems to be great for VB-based programmers.

    14. Re:VS upgrade cycle by h4rr4r · · Score: 2, Insightful

      CPAN installs everything you need, that is why the tool cpan exists. If it takes extra work, you are doing it wrong.

      1. if it does not do what you want, don't use it
      2. just fix the code like you would fix your own, or replace it.
      3. if you don't have source don't use it. No blackboxes that way.
      4. So long as code is all internal who cares.
      5. Use the features you need.
      6. Again if you don't have source you are a fool to use it.
      7. No, this is wrong. This is why every damn windows apps brings along it's own copy of everything. Get a fucking package manager you dolts.
      8. same issue when you have more than 1 developer internal or external.

    15. Re:VS upgrade cycle by QuantumG · · Score: 1

      some random examples of the 2005 to 2008 transition:

      * The entire Tool menu is different, in particular, Lookup Error is gone.. annoys me daily.
      * The solution file format didn't change but they still added the "convert solution" nonsense that means you have to maintain two sln files to maintain backwards compatibility with 2005.. and means people who are using 2008 simply can't supply sln files to 2005 users.. and the vcproj files often need hacking. Why can't they maintain backwards compatibility... it's a text file!
      * The Build Order dialog is completely gone. In 2008 Microsoft decides your build order, no control for you.

      but, primarily, I dislike the forced upgrade of the runtime/compiler with the IDE. Another poster says this is addressed in VS2010.

      --
      How we know is more important than what we know.
    16. Re:VS upgrade cycle by Blakey+Rat · · Score: 3, Insightful

      there's no reason why the IDE release cycle has to be tied to the compiler release cycle..

      There's Intellisense which, you may have heard, is kind of a big deal about Visual Studio. If the language adds new constructs (which is does), and the IDE isn't updated to cope with that, then you end up with useless, or even worse-than-useless, Intellisense.

      What they could do is have the language team produce and release a patch for Intellisense to correspond to their language releases, but then you get into the nasty situation where the language team has to build and QA patches for 3 or 4 versions of Visual Studio-- bugs would creep in, undoubtedly.

    17. Re:VS upgrade cycle by GoatEnigma · · Score: 4, Interesting

      If no one ever reinvented the wheel, we'd all be running around on stone wheels.

    18. Re:VS upgrade cycle by micheas · · Score: 1

      <snip>

      1. Easier for others to follow. If someone else is tracing you code and you get a 3rd party tool, it is either a black box application that you are referencing where you need to break your trace to see what the heck it is for. Or there is source and you go in there and it being code by a different person it is like you are in Wonderland were all coding styles have changed.

      <snip>

      Although, if you use a well known and used library, it can make it easier to find a programmer that knows their way around at least part of your program, and can may get up to speed quickly. Although this applies to frameworks more than libraries, it still holds true to an extent.

      This is more true if you are using opensource libraries.

    19. Re:VS upgrade cycle by Anonymous Coward · · Score: 0

      I have Error Lookup under my tools menu.

    20. Re:VS upgrade cycle by Sarten-X · · Score: 1

      Your statement about CPAN bothers me. But I am a fan of Perl so I digress...

      I agree with most of your points about libraries, but I feel you missed a big one: bugs. My job involves gluing together a few third-party libraries to fit our product. (various Apache stuff, mostly). Now and then, we hit bugs. Not major ones, but significant enough to cause a few hours of lost productivity.

      --
      You do not have a moral or legal right to do absolutely anything you want.
    21. Re:VS upgrade cycle by Com2Kid · · Score: 1

      The solution file format didn't change but they still added the "convert solution" nonsense that means you have to maintain two sln files to maintain backwards compatibility with 2005.. and means people who are using 2008 simply can't supply sln files to 2005 users.. and the vcproj files often need hacking. Why can't they maintain backwards compatibility... it's a text file!

      The solution file is XML in both 2005 and 2008, true, but the capabilities of the file format have increased to support the new solution types of VS2008, amongst probably other reasons.

    22. Re:VS upgrade cycle by QuantumG · · Score: 1

      yes, and there's no reason why they couldn't maintain backwards compatibility... They just didn't try.

      --
      How we know is more important than what we know.
    23. Re:VS upgrade cycle by Anonymous+Brave+Guy · · Score: 1

      One of us misunderstood the earlier posts, and I think it's probably you. The proposal isn't about upgrading the compiler without upgrading the IDE, though I can see why some people would want to do so given VS's history of getting slower and more bloated with each release. In this thread, the proposal is about upgrading the IDE without being forced to upgrade the compiler, which makes sense if you rely on legacy but now non-standard behaviour, or if you're encountering compiler bugs but could work around them with the compiler you've been using prior to the update, to give two real world scenarios.

      --
      If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
    24. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      Then one of us is definitely confused, because VS can be set to use the old compilers. Well, maybe not for all languages, but definitely for C#... I'm working on a .NET 2.0 project in VS2008 right now. (VS2008 was released with .NET 3.5.)

    25. Re:VS upgrade cycle by Anonymous+Brave+Guy · · Score: 1

      FWIW, I was thinking of C++, which has been the cripple of the VS family ever since the first .Net version. It might be possible to set up a custom build process using an older compiler with the newer IDE, but it's been a hassle at best at least up to VS2008.

      --
      If you disagree, post your argument. (-1, Overrated) isn't your personal censorship tool for views you don't like.
    26. Re:VS upgrade cycle by grcumb · · Score: 2, Interesting

      There's Intellisense which, you may have heard, is kind of a big deal about Visual Studio. If the language adds new constructs (which is does), and the IDE isn't updated to cope with that, then you end up with useless, or even worse-than-useless, Intellisense.

      It's interesting to contrast this with bash TAB-completion, which operates in a very similar fashion (albeit in a somewhat different context). The basic capability has been around for a long time, but increasing numbers of utilities are taking advantage of it.

      These days, for example, you can use TAB-completion to see a list of available packages to install based on a given string; you can find the available options and arguments for countless different utilities, even when there are hundreds of available choices.

      All of these improvements have arrived piecemeal, without requiring any particular effort from the bash developers themselves.

      So why should Intellisense language bindings be limited to a particular version of Visual Studio? Surely one should be able add/update/change a language and have the bindings Just Work...?

      I risk inciting a flamewar for saying this, but both vi and emacs can integrate new language/version modes without any fuss or bother. Why would it be an issue for VS?

      --
      Crumb's Corollary: Never bring a knife to a bun fight.
    27. Re:VS upgrade cycle by Onymous+Coward · · Score: 1

      But I for one wish simply to drive my car rather than build it. You gonna have me reinvent the wheel yet again?

    28. Re:VS upgrade cycle by Hognoxious · · Score: 1

      bugs would creep in, undoubtedly

      Bugs? In Microsoft products? How very dare you!

      --
      Confucius say, "Find worm in apple - bad. Find half a worm - worse."
    29. Re:VS upgrade cycle by Victor+Liu · · Score: 1

      From what I can tell, VS2010 is going back to the old model of just including all the DLLs in the app directory. VS2008 was a huge pain with its manifests and SxS assemblies and crap.

    30. Re:VS upgrade cycle by HarrySquatter · · Score: 1

      * The Build Order dialog is completely gone. In 2008 Microsoft decides your build order, no control for you.

      Really? That's funny that I brought up the exact dialog box that you claim doesn't exist.

    31. Re:VS upgrade cycle by n3v · · Score: 1

      Ya, SxS was a [not] nice surprise when I started setting up the testing environment.

    32. Re:VS upgrade cycle by YttriumOxide · · Score: 2, Informative

      most people don't upgrade until necessity forces it on them.

      My "day job" is C#, using Visual Studio, and yes, I'd agree with this in general. Not exactly for the reasons you mention, but close enough to it. My job is basically maintaining and extending an SDK that's handed down from our parent company and then handing it over further to third party development companies (plus a bit of in-house coding ourselves, using the same SDK) and providing code level support for them. The current release of our SDK is entirely .NET 2.0 with VS2005 solution files. We've only just very recently started releasing demo applications for the third parties with VS2008 solution files. It'll be a very long time before we even think about touching VS2010 outside of my "playground" Virtual Machines, since any upgrade we do, we're essentially forcing on several hundred other companies...

      --
      My book about LSD and Self-Discovery
      Also on facebook as: DroppingAcidDaleBewan
    33. Re:VS upgrade cycle by QuantumG · · Score: 1

      See how it says "use the dependencies tab to change the build order" ... no Microsoft, *I* want to change the build order. I had control, you took it away.. why?

      --
      How we know is more important than what we know.
    34. Re:VS upgrade cycle by TrancePhreak · · Score: 1

      Unfamiliar with grabbing the platform SDK are we? Grabbing that is free and can be used on older versions of the IDE.

      --

      -]Phreak Out[-
    35. Re:VS upgrade cycle by NickFortune · · Score: 1

      If no one ever reinvented the wheel, we'd all be running around on stone wheels.

      Furthermore, most of them would be square. Some of them would be triangular. And there'd be all sorts of marketing twonks saying things like "horsepower's cheap: they'll turn round just fine if you put enough power behind them". And "don't go complaining about the bumpy ride. It's not our fault if you're too cheap to pay for decent suspension".

      Reinventing the wheel is allowed. Just make sure your version has more corners than the one you're replacing

      --
      Don't let THEM immanentize the Eschaton!
    36. Re:VS upgrade cycle by Com2Kid · · Score: 2, Insightful

      What do you mean by backwards compatibility?

      Lets say they allow you to open up solutions file unchanged from VS2005. When ever you load a VS2005 solution, any new VS2008 feature that has some aspect of it saved in the solution file would have to be disabled. That also means additional code paths all over the place checking if you have a VS2005 solution loaded up or a VS2008 solution loaded up, and the displaying of appropriate error messages whenever the user tries to use such a disabled feature.

      That is a messy solution, especially given that the most common imagined scenario is for a team of programers working together to all be using the same version of Visual Studio.

      For a product like Word it may be worth it to extend the extra engineering effort to enable backwards compatibility for years on end (IIRC Office 2000 files can interoperate with Office XP and 2003 files A-OK, so two major releases), but for a product used by technical users, one expects that the technical users would rather have other aspects of the product polished more (gated checkins! Holy cripes, TFS finally has gated checkins!) and for developers to just run a one time wizard and convert their project over.

      And actually it is possible to maintain a project across both VS2005 and VS2008. If you need a project to be used by both versions of VS, you only have to bother when changing project references or adding new source files, open the appropriate solution file up in both versions of VS and make the change to the project or the solution.

      Yes it sucks. I've hit this same issue when VS2008 first came out and not everyone on my team was using it, yes it is a PITA. The solution was to tell everyone that they had to just get around to installing VS2008 before the next time they pulled down the latest copy of the code.

      (And yah I see how it could suck if you have multiple teams and some team wants to stay with VS2005 for whatever reason)

      But really, VS2008 supported so many new types of solutions, and in some cases, completely brand new ways to have certain types of solutions organized, that the file format was going to change. Just thinking about it for 10 seconds or so and it seems like maintaining back compat would have involved a UI nightmare, telling users seemingly random features are disabled always sucks, and should be avoided whenever possible.

    37. Re:VS upgrade cycle by LinuxAndLube · · Score: 1

      Lookup Error is gone.. annoys me daily.

      You should try out this new Google search thing. I found the solution to your problem in less than 20 seconds. Makes me wonder what kind of a developer you are, especially considering that the missing menu entry annoys you daily.

    38. Re:VS upgrade cycle by Com2Kid · · Score: 1

      Meh, its almost all MSBuild underneath. I am not a fan of humans editing XML, but if it bothers you that much, go change it. :) Not that hard.

      I had to do this for a few situations as well, where the IDE refused to let me tweak something that needed to be tweaked. One advantage of XML is that at least it is a known evil! (And MS Build is about as readable as I guess one could make an XML based build system that is as flexible as it is...)

    39. Re:VS upgrade cycle by serviscope_minor · · Score: 2, Informative

      It's interesting to contrast this with bash TAB-completion, which operates in a very similar fashion (albeit in a somewhat different context). The basic capability has been around for a long time, but increasing numbers of utilities are taking advantage of it.

      I think you may musunderstand how BASH tab completon works. The completion has nothing to do with the utilities. BASH has programmable completion, and the program usually decides what to do based on the firstr token (i.e. the command name). It will then make finer grained decisions based on the argument before the cursor.

      This is all programmed in BASH, and the utility authors have no input. What has changed recently is that distributions have started shipping with a long and rather complete program for the command history.

      But as you have surmised correctly, the changes have arrived piecemeal. BASH scripts are easy to parse, and all my completion programs make use of only simple text maniuplation to figure out what to do. C++ on the other hand is very much harder to parse.

      Basically to make intellisense work, you need a C++ parser, and C++ is sadly one of the nastiest languages to parse. I would strongly suspect that they upgraded the compiler front end. To go with it, they updated the back end (the fron tnd has new features), the libraries (TR1 benefit from and C++0x libraries require C++0x support), and intellisense.

      By the time they've done that, they may as well upgrade the GUI.

      --
      SJW n. One who posts facts.
    40. Re:VS upgrade cycle by QuantumG · · Score: 1

      'cept for all the guids.

      --
      How we know is more important than what we know.
    41. Re:VS upgrade cycle by weicco · · Score: 1

      Solution files are a mess. I really don't know why they can't be backward compatible. But luckily there aren't many changes between, at least, 2005 and 2008 versions so I managed to write a simple converter which runs 100% transparently. It replaces Visual Studio Version Selector, upgrades/downgrades solution file (plus some other files) and finally calls VS Version Selector to start up the IDE.

      But apart from that "little" problem, I don't see any problem with moving up to VS2010. Just make VS to target .NET version 2.0. I've found VS2010 to be much more productive than 2008, not to mention 2005 which is real PITA compared to either of those. Unfortunately I'll be stuck with 2008 at work for some time now :(

      --
      You don't know what you don't know.
    42. Re:VS upgrade cycle by Anonymous Coward · · Score: 0

      If Microsoft made it easier to use the new IDE without upgrading the compilers, the standard lib, the header install, etc,

      They do, in 2010 we're using 2010 with the 2008 toolchain. Its a simple drop down menu option

    43. Re:VS upgrade cycle by Com2Kid · · Score: 1

      As long as you keep them all straight, nothing bad should happen.... /me ducks head between legs.

    44. Re:VS upgrade cycle by dkleinsc · · Score: 1

      No, actually, the first wheels were probably made of wood, but were completely solid rather than spoked. Stone is a bad choice for constructing wheels: it's a lot heavier than wood, and harder to shape. And while the Sumerians and other folks who thought to build wheels didn't have all the knowledge we have now, they weren't stupid.

      Of course, your point remains valid.

      --
      I am officially gone from /. Long live http://www.soylentnews.com/
    45. Re:VS upgrade cycle by knarf · · Score: 1

      The completion has nothing to do with the utilities. BASH has programmable completion, and the program usually decides what to do based on the firstr token (i.e. the command name). It will then make finer grained decisions based on the argument before the cursor.

      Have a look at the contents of the /etc/bash_completion.d/ directory:

      $ ls /etc/bash_completion.d/
      ant bzr dd gcc imagemagick lisp mkinitrd p4 qemu samba tar xmlwf
      apache2.2-common bzr.simple debconf gcl info lvm mock pbuilder quilt sbcl tcpdump xmms
       
      ...etc, the list does not come through the lame lameness filter.

      Notice those individual tool/command names? Those are completion snippets for the specific command. A tool author can create one of those to tailor bash completion for his/her creation.

      --
      --frank[at]unternet.org
    46. Re:VS upgrade cycle by nacturation · · Score: 1

      He was implying no such thing.

      Most serious Microsoft-oriented shops will do X =/=> All shops who do not do X are not both serious and Microsoft-oriented.

      It does follow logically:

      1. If (serious Microsoft shop) then (will upgrade to VS2010)

      If you've studied logic, you know that given "If A then B" then it logically follows that "If (not B) then (not A)". Thus:

      2. If (won't upgrade to VS2010) then (not a serious Microsoft shop)

      Perhaps it wasn't meant to be read in an if/then format, but if you do read it that way then statements 1 and 2 are logically equivalent.

      --
      Want to improve your Karma? Instead of "Post Anonymously", try the "Post Humously" option.
    47. Re:VS upgrade cycle by RichiH · · Score: 1

      There's Intellisense which, you may have heard, is kind of a big deal about Visual Studio. If the language adds new constructs (which is does), and the IDE isn't updated to cope with that, then you end up with useless, or even worse-than-useless, Intellisense.

      Disclaimer: I never used VS, never will. I may be making wrong assumptions etc pp.

      You seem to imply that the toolset around the compiler needs to newer or exactly as old as the compiler. While there are ways to get around that I'd say that this is painfully obviously the case pretty much everywhere. That would still mean I can use newer tools around an older compiler and stuff works. Which is what GP wanted, imo.

    48. Re:VS upgrade cycle by hclewk · · Score: 1

      You can also do that with Visual Studio, right there in your code.

      /// <summary>
      /// You'll see this in intellisense
      /// </summary>
      /// <param name="x">Intellisense info about what 'x' is</param>
      /// <returns>Intellisense info about what is being returned</returns>
      public string MyFunction(int x)
      {
      //do some stuff
      }

      See the intellisense

      The tricky part comes when introducing new language constructs, not just new classes/functions etc. Try using LINQ in VS 2003. It's just not gonna happen... the IDE doesn't know what to do with it.

    49. Re:VS upgrade cycle by Anonymous Coward · · Score: 0

      Actually, one of the features of 2010 is that you can target old compiler versions (starting with VS2008) with the new IDE.

      One of the primary reasons I don't develop Windows-only (Visual Studio) anymore is because I'd get panic calls because an app had a problem and needed a one-line code fix. But the software couldn't be built anymore because it could only be built via Visual Studio and the current release of Visual Studio was incompatible with the version of VS that the app had been created under.

      Java has its warts, no question. But at least I can still build stuff from 10 years ago without having to uninstall my current IDE, dig out an older IDE, install and patch it (assuming the CD's still readable), and if I'm lucky - not discover that it only runs under Windows 98/NT4.

    50. Re:VS upgrade cycle by somersault · · Score: 1

      It wasn't meant to be boolean logic, he said "most". So you'd need some kind of probability in there.

      --
      which is totally what she said
    51. Re:VS upgrade cycle by vivian · · Score: 2, Interesting

      My "day job" is C#

      Sorry to hear that. Mine used to be too - then I started using QT and QtCreator for my main project. Bliss! Finally I am able to easily write apps that will run on multiple platforms, but still has a rich library of UI controls ( widgets) and plenty of abstract data types.
      I really cant see me going back to C# - I just don't see a need for it any more, now it is so easy to write GUI intensive apps in C++. I have been working on my current project for a couple of years in C#, but it has only taken me a few months to get a C++/QT equivalent up and running - I was able to even ditch a lot of the code I had written in C# (eg. a docking panel interface, a MVC for treeviews, etc) because QT already provided those. In addition I have been able to completely ditch the third party control I was using for opengl too, since again, there's already an opengl widget in QT. (I'm writing a sort of Quantity surveying system / scriptable 3d CAD / quoting database thing for a construction company)
      Only down side I had initially to moving to QT was getting my head around how QT form designs and widget layouts work compared to how forms work in C# (ie. using anchoring/docking for sizable forms), but I have pretty much nailed that now.

      My biggest bitch with C# and Visual Studio Professional 2005 was the Complete lack of a profiler - I mean - WTF! I'm not forking out $15,000 for the team edition to get a profiler.

    52. Re:VS upgrade cycle by fbjon · · Score: 1

      This may be a dumb question, I have no experience with VS, but why is it called a solution file?

      --
      True confidence comes not from realising you are as good as your peers, but that your peers are as bad as you are.
    53. Re:VS upgrade cycle by sproketboy · · Score: 1

      Odd. Since IntellIJ works fine with Java 6 even though it internally uses Java 5 (at least in my version). BTW Intellisense in Visual Studio is a bit of a misnomer since it's largely unusable without Resharper. A better name would be Retardsense.

    54. Re:VS upgrade cycle by bondsbw · · Score: 2, Informative

      But the software couldn't be built anymore because it could only be built via Visual Studio and the current release of Visual Studio was incompatible with the version of VS that the app had been created under.

      That hasn't been the case in many years. Every .NET version of Visual Studio (2003, 2005, 2008, and 2010) are backwards compatible with the previous project files and solutions from prior .NET versions. Since I don't use VS for C++ or VB, I can't say whether compatibility goes back further.

      As far as the language and libraries go, Microsoft makes a point to keep things compatible unless there was a major bug with an API. C# 4.0 will compile C# 1.0 source code, and the libraries will still work, assuming you used them correctly and are not dependent on bugs in 1.0/1.1. (One major exception... ASP.NET 1.0/1.1 code is somewhat incompatible with later versions. I call this a feature, considering that the newer versions more strictly conform with the XHTML standard.)

      .NET 3.0+ features like WPF are not available on certain older operating systems, but that's a different subject. The 3.0 features are just additional libraries that run on CLR 2.0. If you don't use those libraries, compile your code for .NET 2.0 and you can use it on the older operating systems. 2.0 compilation has been provided since its introduction in VS 2005.

      --
      All my liberal friends think I'm a conservative, all my conservative friends think I'm a liberal.
    55. Re:VS upgrade cycle by godefroi · · Score: 1

      I've never used crutchsharper, and I've been writing .net code since the original betas. In VS2010 intellisense is even better.

      --
      Karma: Poor (Mostly affected by lame karma-joke sigs)
    56. Re:VS upgrade cycle by godefroi · · Score: 1

      In my experience, commercial "component" vendors produce some of the worst software in existence. If my company produced that kind of stuff, we'd have been out of business a decade ago.

      I think that means I agree with you.

      --
      Karma: Poor (Mostly affected by lame karma-joke sigs)
    57. Re:VS upgrade cycle by godefroi · · Score: 1

      That hasn't been the case in many years. Every .NET version of Visual Studio (2003, 2005, 2008, and 2010) are backwards compatible with the previous project files and solutions from prior .NET versions.

      That's not really strictly true. Each new version of VS can CONVERT earlier projects and solutions to a current format, but it cannot directly load the earlier file formats.

      VS2008 introduced the ability to target some earlier "versions" of the .NET Framework (though it only targets frameworks based on .NET 2.0 (2.0, 3.0, 3.5 are all the 2.0 Framework, though 3.5 introduced a new compiler).

      VS2010 can target the 2.0, 3.0, 3.5, and 4.0 Frameworks. Between these 4 "versions" there are actually only two runtimes, and three compilers.

      --
      Karma: Poor (Mostly affected by lame karma-joke sigs)
    58. Re:VS upgrade cycle by godefroi · · Score: 1

      A "solution" is a grouping of "projects".

      Interestingly enough, a "project" (since VS2005) is really an XML-based MSBuild "script" (extremely similar to an Ant script, if you're Java-ish, or a makefile if you're into that sort of thing). A "solution" is still text-based, though MSBuild has the brains to convert a solution into a more project-style thing when it's asked to build one.

      --
      Karma: Poor (Mostly affected by lame karma-joke sigs)
    59. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      Notice those individual tool/command names? Those are completion snippets for the specific command. A tool author can create one of those to tailor bash completion for his/her creation.

      1) You can do that with Intellisense, which you seem to be completely ignorant of and so I have no idea why you're continuing this debate.

      2) The problem is when the language *syntax* changes. When .NET added getters and setters, if Intellisense wasn't aware of those, it would have no way of doing any sort of auto-completion-- in fact, to Intellisense, it would just look like you were typing syntax errors.

      The point is that Intellisense has to actually be able to parse the language, and since all of the languages included with Visual Studio change regularly, Intellisense has to be updated regularly as well. Even C++ has been updated in every release.

      Another side effect is that languages that are nearly-impossible to parse (for example, JavaScript) don't get much Intellisense help.

    60. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      That would still mean I can use newer tools around an older compiler and stuff works. Which is what GP wanted, imo.

      Well, like I said in another reply, then somebody's obviously confused, because you can already do that in Visual Studio and have been able to for ages. It's not necessarily trivially easy to configure, but it's definitely possible, and VS ships with everything needed.

    61. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      BTW Intellisense in Visual Studio is a bit of a misnomer since it's largely unusable without Resharper.

      Resharper got a ton of sales when Intellisense wasn't all that great. But you know what? Go through Resharper's "feature list" now, and you'll just be saying: "Intellisense already does that. It already does that. Already does that." Resharper has very little that in it that isn't already in Visual Studio 2008, I imagine it has even less that's not in 2010.

      Seriously, go through Resharper's feature list, and tell me what it's doing that Intellisense doesn't already do.

      You're just buying it out of habit at this point, not out of need.

    62. Re:VS upgrade cycle by RichiH · · Score: 1

      Ah, k. Thanks for that clarification. As I said, I never used VS and based my posting on a lot of assumptions.

    63. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      Yeah. Unfortunately, the worse assumption is always: "people who post on Slashdot know what they're talking about." (Including me, of course.)

      That said, apparently, VS2010 makes it much easier to target older compiler versions. So the feature isn't "it can do it", but "it's easy to do it." I haven't tried 2010 yet, but I think I'll download the Express version right now, now that I'm thinking about it.

    64. Re:VS upgrade cycle by HarrySquatter · · Score: 1

      So you switch tabs and use that to change the build order how ever you want. Wow, so hard.

    65. Re:VS upgrade cycle by YttriumOxide · · Score: 1

      Sorry to hear that. Mine used to be too - then I started using QT and QtCreator for my main project. Bliss!

      I'm honestly jealous... while I don't dislike C#, I'm definitely not a huge fan of it. I started my "serious" programming life with C++ (although dabbled in many other languages before that), so I'd love to go back if I could. Sadly, for my day job at least, I'm at the mercy of the parent company on this...

      Finally I am able to easily write apps that will run on multiple platforms, but still has a rich library of UI controls ( widgets) and plenty of abstract data types.

      About 75% of everything I do for work has to be cross platform under mono. The "core" of all of my apps is therefore always specifically designed with this in mind. I DO re-write all of the GUI though for any app that has one (using WinForms on Windows, GTK# on Linux, and Cocoa# on MacOS if I have to target that also). That really is a complete PITA, but if kept in mind from the start it generally only adds an extra 2 to 3 days on any project I do. I could be lazy and just use GTK# on all three, but it looks pretty ugly on both MacOS and Windows, so I prefer to take the time for a "native" look and feel.

      Only down side I had initially to moving to QT was getting my head around how QT form designs and widget layouts work compared to how forms work in C#

      I haven't looked at QT much, but the small amount I did, it sort of reminded me of a really hyped up version of "MUI" on AmigaOS from the layout perspective. If my impression was correct, I probably wouldn't have too many migration pains were I try.

      I'm not forking out $15,000 for the team edition to get a profiler.

      One advantage of working for a huge multinational I guess... For Microsoft stuff at least, I just log on to MSDN and download pretty much anything I want/need (including the "Team Edition"). I certainly can't imagine actually PAYING for Visual Studio (or Windows, or Office, etc etc)...

      --
      My book about LSD and Self-Discovery
      Also on facebook as: DroppingAcidDaleBewan
    66. Re:VS upgrade cycle by sproketboy · · Score: 1

      Just off the top of my head.

      * Auto import of namespace references
      * Code insight - know if something is broken without having to compile everything (BTW something that Java IDEs have had since the 90s)
      * Proper ability to re-factor code where XML is involved.

    67. Re:VS upgrade cycle by Blakey+Rat · · Score: 1

      Ok, the first and third one I'll give you.

      But unless there's more to "code insight" other than what you just described, VS already does that-- has for ages.

    68. Re:VS upgrade cycle by Anonymous Coward · · Score: 0

      "My "day job" is C#

      Sorry to hear that."

      Yeah, how terribly horrid that must be. I mean, imagine the horror of having a skill that would most likely allow you to find another job and stuff, lol.

    69. Re:VS upgrade cycle by rdnetto · · Score: 1

      If it's just a one line change, use Notepad and MsBuild. MsBuild is part of the .NET runtime, so you should have no trouble finding a copy.

      --
      Most human behaviour can be explained in terms of identity.
    70. Re:VS upgrade cycle by orasio · · Score: 1

      most serious Microsoft-oriented shops will upgrade to Visual Studio 2010 and never look back

      Of course, implying that you're not a serious Microsoft-oriented shop if you don't upgrade.

      Or implying that VS 2008 is a POS, and any sensible person will trade it for anything that promises to be better.
      I'm just giving you another possibility. I don't know VS2008 _at_all_. In fact, I had to work through 2008 with VS2005 because the shop would not authorize 2008. My opinion at the time was that VS2005 was a POS, didn't have a responsive UI and lacked very mainstream editing capabilities. My coworkers kept saying that I was right, 2005 was a POS, but that the next version was going to be great. I step exactly every five years in the VS world, and the situation is always like that. This version is crap, but the next one is going to be great and non frustrating at all.

  2. Whew! by Anonymous Coward · · Score: 2, Funny

    proper support for time after the Y2038 barrier

    Good to know that my time machine written in PERL will no longer malfunction due to improperly handled timestamps! Now to test this baby out. This knob here, this button here, and... %^$%^$%^$%^$%^ NO CARRIER

    1. Re:Whew! by biryokumaru · · Score: 1

      Dude, that's where he went in his time machine. Duh.

      --
      When you're afraid to download music illegally in your own home, then the terrorists have won!
    2. Re:Whew! by badboy_tw2002 · · Score: 1

      I'm confused, are you from the past and your time machine connects to the future on a modem?

    3. Re:Whew! by suso · · Score: 1

      Oh wow, I didn't know the Time::Warp module did that. I always ignored that module.

      Perl has a module for everything.

    4. Re:Whew! by t-er-eyma · · Score: 1

      Er... not trying to spoil your joke... but... the issue is not with Perl, but with most 32bit Linux machines, simply because time_t is a 32bit number and it overflows after 2038. That's a real problem that is out there and looks worse then the Y2K issue. But maybe you're just hoping there will be no 32bit binaries in use in 2038, in which case you don't have to worry. But notice that this problem currently affects any 32bit binary that uses the system functions to handle date and time.

    5. Re:Whew! by __aasqbs9791 · · Score: 1

      Many of the people reading this site won't be here in 2038. Some due to age related causes, and many due to other causes. *

      * Brought to you by your favorite anti-depressants. Now back to your regularly scheduled program.

    6. Re:Whew! by suso · · Score: 1

      Too bad perl didn't solve the 2006 problem. You know, the fact that it's been irrelevant for 4 years.

      Why? Because its old? You still use Linux, right? If you run a Mac, that runs Unix. Both those OSes include Perl by default.

      By the way, you're still on Slashdot, that's old too and also runs on Perl. Don't be an idiot.

    7. Re:Whew! by Anonymous Coward · · Score: 0

      Remember that 2038 comes about in around about 27.75 years. 27.75 years ago was mid-1982. I was still doing my computing with a DECWriter II teletype spitting out reams of green-bar paper connected via a 300 baud breadbox modem to a PDP-11/44.

      My point is 27.75 years is a long time in this industry. 64-bit hardware is becoming more and more ubiquitous, the OS's are starting to follow, and apps are as well. Will there still be 32-bit apps around in 2038? Yeah, I'm pretty certain of it. However, I think most of them will have been phased out, and what remains will get fixed with urgency, but nothing like the Y2K problem.

    8. Re:Whew! by mr_mischief · · Score: 1

      So they can use a double word for their time_t rather than a word. Problem solved.

    9. Re:Whew! by TheLink · · Score: 1

      Perl is often very useful because it's installed by default on many unix and unixlike operating systems, and it is easier to do stuff in it than do stuff using shell scripts for all those operating systems.

      Something newer like Python or Ruby won't be pre-installed on as many operating systems.

      --
    10. Re:Whew! by MacGyver2210 · · Score: 1

      I don't know where you learned your bitmath, but 32-bits *is* a DWORD

      I think the logical solution which should've presented itself after the whole Y2K scare thing, would be to use a time-since-Jan-1st number for the day/month and a separate integer for the year. Even if it's an unsigned short, it can still go somewhere around 16,500 years before we're going to run out of digits.

      --
      If the only way you can accept an assertion is by faith, then you are conceding that it can't be taken on its own merits
    11. Re:Whew! by Anonymous Coward · · Score: 0

      A "word" is the size of the data used by the processor. A word on a 32-bit process is 32 bits, and a word on a 64-bit processor is 64 bits. Microsoft (and others, but most notably MS) has kept calling 16-bit numbers "words" in their libraries because they are lazy. These names are inaccurate for todays (and yesterday's) computers.

      Also you seem to be confused about the point of keeping time this way. The whole point of unix time is that you have ONE piece of data to compare and pass around which represents the COMPLETE date down to the second. Your "solution" doesn't respect this, which makes it useless. If you're going to start using multiple values, you might as well just represent each part of the date/time separately.

    12. Re:Whew! by jonadab · · Score: 1

      Actually, we solved that back in the nineties when we secretly rewrote about a quarter of the world's software in Perl and never told anyone. So now everybody's been using software for years that requires Perl, and they never even realized. Think of it as security through surreptitious ubiquity.

      --
      Cut that out, or I will ship you to Norilsk in a box.
    13. Re:Whew! by Anonymous Coward · · Score: 0

      I think you only turned up the bubbles on your hot tub.

  3. Stereotyping? by kickme_hax0r · · Score: 2, Insightful

    So almost every developer either: a) Uses Erlang and a Mac b) Uses Perl or c) Uses VS2010? Is it just me, it would there still be a fair chunk of developers not included?

    1. Re:Stereotyping? by Anonymous Coward · · Score: 0

      It's just you.

    2. Re:Stereotyping? by dudpixel · · Score: 1

      yes, you are the fair chunk of developers that are not included :P

      --
      This seemed like a reasonable sig at the time.
    3. Re:Stereotyping? by FooAtWFU · · Score: 3, Informative

      Well, almost every Unix developer uses Perl for glue code, and almost every Microsoft developer will use VS2010, and if you're programming a Mac, I don't see how you could be sufficiently non-erudite to use anything but Erlang.

      --
      The World Wide Web is dying. Soon, we shall have only the Internet.
    4. Re:Stereotyping? by Anonymous Coward · · Score: 0

      Shut up, dude. This is still one of the better ones to come out in a while. Look at all the other posts recently. And I HATE Perl.

    5. Re:Stereotyping? by Haxamanish · · Score: 2, Funny

      I really hoped there would be something in this thread for Prolog programmers who write their time-critical code in assembler, but I was disappointed again.

    6. Re:Stereotyping? by julesh · · Score: 1

      Well, almost every Unix developer uses Perl for glue code

      Really? When I'm wearing my Unix dev hat, I usually use shell for such functions. I figure if it's good enough for the system's init scripts, it's good enough for me...

      and almost every Microsoft developer will use VS2010

      And in my Windows hat, I'm still using VS2005. Not sure if I'll upgrade or not; from what I hear I'll need a new workstation first.

      and if you're programming a Mac, I don't see how you could be sufficiently non-erudite to use anything but Erlang.

      While I don't program Macs, I have to express shock at the idea that a true Mac developer would consider anything other than Objective C.

    7. Re:Stereotyping? by luke_z3 · · Score: 1

      I, too, was hoping for an update on OpenCOBOL

    8. Re:Stereotyping? by FooAtWFU · · Score: 1

      That was a joke, son. Mostly about the Erlang...
      not sure why it wasn't marked funny, though.

      --
      The World Wide Web is dying. Soon, we shall have only the Internet.
    9. Re:Stereotyping? by Anonymous Coward · · Score: 0

      I actually did exactly that once. Ah, to get Turbo Prolog running again.

    10. Re:Stereotyping? by Xest · · Score: 1

      But what about Java developers?

  4. So... by FlyingBishop · · Score: 5, Insightful

    We've started churning out pointless stories all day and then cramming four actual news posts into a single thread?

    1. Re:So... by crazycheetah · · Score: 4, Funny

      Welcome to Slashdot!

    2. Re:So... by dudpixel · · Score: 1

      where have you been?

      --
      This seemed like a reasonable sig at the time.
    3. Re:So... by Blakey+Rat · · Score: 2, Insightful

      At the very least, it could have been 4 actual entries so that the comments don't just become a huge clusterfuck.

      But... nobody running Slashdot gives a fuck. They don't even pretend, really.

    4. Re:So... by FlyingBishop · · Score: 1

      Oh, I've been around, and I don't really mind the pointless shit (you can just ignore it after all) but this is all stuff that's worth discussing... in separate threads.

    5. Re:So... by Taagehornet · · Score: 1

      I guess the space is needed for all the Apple... erhm... news that's been clogging the tubes lately.

    6. Re:So... by Anonymous Coward · · Score: 0

      But... nobody running Slashdot gives a fuck. They don't even pretend, really.

      Pretty much why I stopped visiting when I found reddit. Yo, editors are so last century, crowdsurfing, baby! Democracy of news aggregation!

      Then I realised that democracy is kind of shitty when the majority of the voters are chantards who mostly vote up stories about a site where people parade their dicks on webcam or terrible comics about a man who apparently can't figure out how to finish pronouncing a swear word.

      So now I'm back on slashdot, and have found a new appreciation for even the dupe-ridden, grammar-clusterfucked, link-broken, (etc, etc) "editors" here.

    7. Re:So... by tehcyder · · Score: 1

      Look on the bright side, they're making it harder to post offtopic, which is great news for me as I would like to start a debate about ferret racing and it just never seems to come up here.

      --
      To have a right to do a thing is not at all the same as to be right in doing it
  5. IDEs by JohnFluxx · · Score: 1

    I wonder what other developers think of IDEs.

    I really like the intelliSense etc as an idea. But in practice the whole IDE thing gets in my way.

    IntelliSense does let me write code faster, but I find that I don't spend most of the day actually typing in code as fast as I can. I spend a large amount of time using git (revision control tool) on the command line, or debugging and profiling the app, using strace, nm and other command line tools, and so on. The IDE really gets in my way most of the time.

    If I switch between branches or revisions (which I do often - git is so wonderful) then IDEs tend to go haywire as hundreds of files suddenly change, and then change again and again..

    So, I stick to vim. I know there's ctags etc but I don't even bother with that. After a while I tend to just remember most of the API anyway.

    1. Re:IDEs by badboy_tw2002 · · Score: 4, Interesting

      When your project size starts getting large and the number of classes/functions/types/etc starts heading to the thousands its pretty nice to have something that will quickly show you the organization of the code base and help you find things faster. Stuff like "I'm in a source file, open the corresponding header" or "show me all the places that call this function" or "rename this function everywhere it was used" or even "let me browse through the 10 versions of the function to see the right one without having to load the header file and stare at it". Also, when there's tight checkout integration its nice to click on another file, check it out, etc. without having to drop to the command line or move to something else. That's not to say that the command line isn't useful - I still find it easier to sometimes run makes or grep or whatever so there's always one handy, but personally I get a lot more done than with a plain old editor.

    2. Re:IDEs by jellomizer · · Score: 3, Interesting

      I use IDE as a tool.

      The problem is some people use it as a crutch... "If the IDE doesn't handle it then it cant be done" mindset

      But it is a tool to keep in your cap. I know Visual Studio 2008 is good at showing me methods and properties I can access from the variables which is handy to let you know what is going on in a datatype/class you don't use much.
      However sometime I will take the file out of Visual Studios to do some additional coding because Visual Studios Expects you to code in Top Down in this order...
      IF (x == y)
      {
                  msgbox("hello");
      }

      However real life has it more like this...

      msgbox("hello");

      run test...

      Up arrow Return
      if (x == y) {
      return down arrow to go past the msgbox command
      }

      For this case the IDE is a pain because it will try to close my if statments {} which if I am not paying attention it will give me an extra } that I need to dig around and find when I get a compile error.

      --
      If something is so important that you feel the need to post it on the internet... It probably isn't that important.
    3. Re:IDEs by QuantumG · · Score: 1

      agreed. A good IDE is superior to grep.. and that's the right metric. That said, often VS fails so hard at being better than grep that I reach for the built-in grep (it's called Find In Files). If anyone wants to try beating Microsoft at making a better IDE than VS, study the way people use Find In Files and make it better.. for example, if I they added a dropdown on the Find In Files which let me select "function calls", or "templates" or "class definitions", that'd be great.

      --
      How we know is more important than what we know.
    4. Re:IDEs by Anonymous Coward · · Score: 0

      Eclipse is a better IDE than VS.

      And, yeah, I've used both. I think VS is a miserable chunk of shit, and that when people call it the best in the industry, they're either lying or misguided.

      Still, I'm more of a vim/minibufexplorer/ctags kinda guy.

    5. Re:IDEs by QuantumG · · Score: 4, Interesting

      Eclipse might be better than VS, but I've never been able to get it to run fast enough to be usable.

      Last time I installed it the person advocating it to me looked over my shoulder and said "yeah, I think you need to upgrade your video drivers".

      --
      How we know is more important than what we know.
    6. Re:IDEs by abigor · · Score: 4, Interesting

      I really hate Eclipse actually. I find it buggy, slow, and non-intuitive in a lot of ways. For Java stuff, IntelliJ IDEA is really great. For non-MS C and C++...I'm not really sure anymore. I mostly end up using Vim and the command line.

    7. Re:IDEs by binarylarry · · Score: 1

      You should try a recent release. Eclipse used to be very slow on Linux until a few releases ago, due issues in the GTK+ SWT bindings.

      It's always been very snappy on Windows.

      Also don't use any of the natively compiled versions that some distros ship, they're much slower than standard Eclipse. Always get Eclipse from Eclipse.org, because distro packaged versions have historically been quite out of date (although they're working keeping things more recent).

      --
      Mod me down, my New Earth Global Warmingist friends!
    8. Re:IDEs by Anonymous Coward · · Score: 0

      That's funny, because I've used both Eclipse and VS, and I think Eclipse is a miserable chunk of shit, and that VS really is the best in the industry. Interesting how we can both make up arguments without backing them up with anything concrete, isn't it, guy?

    9. Re:IDEs by ClosedSource · · Score: 1

      My experience with Eclipse is that it's an OK IDE for Java, but it handles C++ as if it were designed in 1995. Well, the other 1995 products were actually easier to configure.

    10. Re:IDEs by ClosedSource · · Score: 2, Funny

      "It's always been very snappy on Windows"

      Hmm. And I always thought that "snappy" implied fast and responsive. I guess not.

    11. Re:IDEs by FriendlyLurker · · Score: 1

      I really hate Eclipse actually. I find it buggy, slow, and non-intuitive in a lot of ways.

      My experience as well - I was shocked actually. Recently had to use that latest greatest "Eclipse for a Java" with a legacy Java project for the first time - the disconnect between "WorkSpaces" and "Projects" drive me batty after a while. Got an existing java project with eclipse project file? Forget trying to just open it - Nooo you have to jump through all sorts of non intuitive loops and importing processes just to load the damn thing into the IDE. Documentation ambiguous, scarce and/or non-existent for the problems and error messages Eclipse keeps throwing in your way during that importation process. And all that just to load the damn project - I am not looking forward to developing under such a non-intuitive environment - IDE's are supposed to make mundane tasks easier, not complicate them.

    12. Re:IDEs by TrancePhreak · · Score: 1
      My experience is that most people that don't use IDEs just don't know how to use them.

      If I switch between branches or revisions (which I do often - git is so wonderful) then IDEs tend to go haywire as hundreds of files suddenly change, and then change again and again..

      This for instance is not something you do while the IDE is open. Debugging is also another area where the current crop of IDE's will be better. If you use STL or something similar it can show you the raw types or you can view them in a more meaningful manner, like being able to read a map or a vector as non-complicated data types.

      --

      -]Phreak Out[-
    13. Re:IDEs by JohnFluxx · · Score: 2, Interesting

      > This for instance is not something you do while the IDE is open

      Then that pretty much rules using an IDE. It's crazy to imagine opening and closing an IDE every few minutes just because I want to backport my fixes to a different branch, or bisect to find a bug, etc.

      Even plain old GDB lets you view raw types just by writing macros to display things however you want. Hell, it even has python scripting support, so you can go crazy with it.

    14. Re:IDEs by JohnFluxx · · Score: 2, Interesting

      grep is an interesting point actually.

      Okay, so the IDE lets you search all files. Just like doing:

          git grep someword

      How do I now use the IDE to search all the files.. in all of time? e.g. I remember that the code once had a certain string somewhere, but it seems to have been deleted. How do I find that?
      With git it's a single command.

    15. Re:IDEs by dkf · · Score: 2, Interesting

      Eclipse might be better than VS, but I've never been able to get it to run fast enough to be usable.

      I used to have a lot of problems with that, and then I moved it to a machine with more memory and the problems went away. IOW, it's a bloaty hog but otherwise OK.

      --
      "Little does he know, but there is no 'I' in 'Idiot'!"
    16. Re:IDEs by Anonymous Coward · · Score: 0

      Highlight code, right click, "surround with".

    17. Re:IDEs by fredrik70 · · Score: 2, Interesting

      hear, hear, well, think Eclipse is alright, decent solid IDE once you get to know it. Jetbrains IntelliJ IDEA though rocks everything, best IDE ever.
      Coming back to VS, the best part with it is that you can install the Jetbrains Resharper, which makes VS behave (almost) like IntelliJ!!

      --
      if (!signature) { throw std::runtime_error("No sig!"); }
    18. Re:IDEs by doogledog · · Score: 1

      You could always, y'know, just turn that feature off.
      I love Visual Studio, but I always spend a minute or two configuring a fresh install to how I want it. However I like the fact that I have those options available to me.

      You are correct though, give some coders notepad/emacs/vi/cat and a compiler rather than their usual fully-featured IDE and they're completely lost. But one thing I've noticed (especially recently, working with EEs who often write firmware but don't consider themselves developers) is that not having the time/experience to completely memorise the entire set of library functions available doesn't make someone less of a skilled developer.

    19. Re:IDEs by moonbender · · Score: 2, Informative

      I don't know about "just opening it", but you can simply select Import -- Existing Project into Workspace and that's that. This has never been an issue for me with all kinds of projects (some of them rather large). A workspace is just a set of projects; I think many IDEs have a concept like that. If you don't want to bother with it, you can simply have one workspace for all projects with no real downside, that's what I do at home.

      That said, setting up the development environment for a legacy project can be anything but mundane. You need to have all the dependencies available, deal with version changes, have the compiler settings in place. Project files help a lot with that.

      --
      Switch back to Slashdot's D1 system.
    20. Re:IDEs by moonbender · · Score: 1

      What, you mean with the mouse? Are you kidding? Well, I assume there's a way to invoke that with the keyboard, as well. I'm usually a pretty mouse heavy user, but not when I'm coding.

      --
      Switch back to Slashdot's D1 system.
    21. Re:IDEs by moonbender · · Score: 1

      There's really no need to close the IDE. I often use Eclipse, and so far it has always handled things well when I tell it to "Refresh" the project. Sure, it needs to regenerate some static source analysis data, but that's not easily avoidable. Of course, if you interface with the revision control system from within the IDE, it's even easier. Dunno how good the git integration is for Eclipse and other IDEs, though.

      --
      Switch back to Slashdot's D1 system.
    22. Re:IDEs by Anonymous Coward · · Score: 0

      My experience of Eclipse has been that it is slow, buggy, and generally horribly shit piece of software that embodies everything that is bad about open source (I like OSS in general, btw). And yes, I have used it recently, in numerous 'professional' contexts.

      As for Visual Studio, I can't speak for 2010, but up till now its been an ok IDE at best. Once you install Resharper however, nothing comes close, I'm absurdly productive in it.

    23. Re:IDEs by binarylarry · · Score: 1

      That's exactly what it means and what Eclipse is today on any platform (except Mac... but it's because Mac UI is laggy, not Eclipse).

      --
      Mod me down, my New Earth Global Warmingist friends!
    24. Re:IDEs by Anonymous Coward · · Score: 0

      For this case the IDE is a pain because it will try to close my if statments {} which if I am not paying attention it will give me an extra } that I need to dig around and find when I get a compile error.

      First, you can almost certainly turn that off. And second, how did you get Visual Studio to do that to begin with? Out of the box, mine doesn't insert anything.

    25. Re:IDEs by jisatsusha · · Score: 1

      There is. Select with the keyboard (I doubt I have to tell you how, but hold down shift when you move the input cursor), then press ctrl+k, s. virtually everything in VS has a keyboard shortcut.

    26. Re:IDEs by pavon · · Score: 1

      No, it's not. When I type and intermittently have to wait upto 3 seconds for the text to appear on the screen, it is not snappy.

    27. Re:IDEs by Anonymous Coward · · Score: 0

      Eclipse on the Mac is quite snappy, as is the overall Mac UI. Your perceptions of things may have been formed on earlier versions of the OS and/or hardware. It's no more accurate now than the impression that Windows BSODs every couple hours.

    28. Re:IDEs by godefroi · · Score: 1

      You know you can turn that off, right?

      --
      Karma: Poor (Mostly affected by lame karma-joke sigs)
    29. Re:IDEs by Blakey+Rat · · Score: 1

      > This for instance is not something you do while the IDE is open

      Then that pretty much rules using an IDE. It's crazy to imagine opening and closing an IDE every few minutes just because I want to backport my fixes to a different branch, or bisect to find a bug, etc.

      Well, normally you'd use a version control system that's integrated into the IDE. Then you wouldn't have to close jack.

      But seriously, you do those operations "every few minutes?" I find that very hard to believe.

    30. Re:IDEs by binarylarry · · Score: 1

      That delay is configurable. If you want instant context assistance, set the delay to 0 in the preferences.

      This is something a lot of newbie developers seem to ask for but believe me, you do not want this. But there's no better way to learn.... set the popup delay to 0 and bask in the "fastness."

      For the JDT tools its in Window->Preferences->Java->Editor-Content Assist->Auto Activation Delay

      --
      Mod me down, my New Earth Global Warmingist friends!
    31. Re:IDEs by pavon · · Score: 1

      I don't want instant context assistance, what I want is for the context assistance, and whatever else is going on in the background, to not delay my normal typing.

      Anyway, I don't have Eclipse installed at the moment. I might try it out again next time I start a new java project at home to see if it has improved, but it was intolerably slow when I tried it a year ago.

    32. Re:IDEs by julesh · · Score: 1

      If I switch between branches or revisions (which I do often - git is so wonderful) then IDEs tend to go haywire as hundreds of files suddenly change, and then change again and again..

      This is why you should be using a version control system that integrates with your IDE rather than a command line one; I use AnkhSVN when working with visual studio, and it copes with such things without a huge issue. Sure, you'll need to drop to command line occasionally, but most day to day operations can then be handled by the IDE without running the risk of confusing it.

    33. Re:IDEs by binarylarry · · Score: 1

      I use a Mac on a weekly basis. There's a noticeable lag due to the composition going on to create the nice Cocoa aesthetics. The same thing shows up on Ubuntu when you have effects enabled.

      Great looks but it sacrifices responsiveness.

      --
      Mod me down, my New Earth Global Warmingist friends!
    34. Re:IDEs by julesh · · Score: 1

      How do I now use the IDE to search all the files.. in all of time? e.g. I remember that the code once had a certain string somewhere, but it seems to have been deleted. How do I find that?
      With git it's a single command.

      I have no experience with git, but with an appropriate plugin VS exposes pretty much all the standard subversion commands for me. Unfortunately, subversion lacks that particular facility, so I can't be *sure* it would work, but I imagine any appropriate git plugin for visual studio will provide it.

    35. Re:IDEs by julesh · · Score: 1

      Eclipse might be better than VS, but I've never been able to get it to run fast enough to be usable.

      Strangely, I have the opposite issue. Eclipse runs just fine on my machine (Celeron D 2.66GHz, 1.5GB) but VS2005 + Resharper (because I can't live without automated refactoring) is a total hog. Intellisense often takes multiple seconds to open, startup times are annoyingly long and I often find my machine is lagging and open up the task manager and find devenv.exe is randomly using about 30% of my CPU time for no apparent reason.

    36. Re:IDEs by badboy_tw2002 · · Score: 1

      There are some patches out there for 2005 to help out with Intellisense, but it still pretty much sucked even with that. 2008 makes it unnoticeable, but it still has correctness problems. Visual Assist X makes everything work. Haven't tried 2010, but it looks like they rewrote it, so maybe this time they got it right.

    37. Re:IDEs by badboy_tw2002 · · Score: 1

      Yeah, its pretty much up to your version control, not the IDE (though the IDE version control should expose it). With the exception of MS version control, the plugins are written by the version control devs, so quality varies from ok to ghastly in my experience. (Not that MS version control doesn't suck big time).

    38. Re:IDEs by ClosedSource · · Score: 1

      My observation is that people claim that Java has improved and and is finally competitive with the speed of C or C++ applications.

      The only problem is they've been making the same claim every 2 years since Java was introduced.

    39. Re:IDEs by ClosedSource · · Score: 1

      That's a lot of menu navigation. Why don't they just have a top-level checkbox with the label "Make Performance Suck"?

    40. Re:IDEs by ndogg · · Score: 1

      I really like MonoDevelop. I'm not certain about its Java support, but it does pretty well with C/C++.

      --
      // file: mice.h
      #include "frickin_lasers.h"
    41. Re:IDEs by Anonymous Coward · · Score: 0

      SlickEdit 2009 has every one of these features! Plus it frees you from being platform locked to Windows :) www.slickedit.com

    42. Re:IDEs by rdnetto · · Score: 1

      For C++ I quite like Qt Creator - it's open source, cross platform and the libraries are about as capable as .NET.

      --
      Most human behaviour can be explained in terms of identity.
  6. as far as I can remember ... by jobst · · Score: 0, Flamebait

    Perl had a 5 in front .. in fact since 1994 I think.

    --
    to code or not to code, that is the question.
    1. Re:as far as I can remember ... by rubycodez · · Score: 1, Offtopic

      The Perl 6 Design Team had a 6 in the middle, since 2000, and on and on it goes................

      http://use.perl.org/~chromatic/journal/40296

    2. Re:as far as I can remember ... by rubycodez · · Score: 1

      hah, some fanboy mods didn't like that of talk. But the point is well taken, Perl 6 has failed because of too much feature inclusion and navel-gazing, it's like a 2000 lbs. swiss army life that includes kitchen sink, porta-poddy, and arc welder. Time to put a bullet into the head of that Frankenstein and start over, make a simple, unified clean OO and AO language that is natural extension of Perl 5 but without the kludgy constructs.

    3. Re:as far as I can remember ... by chromatic · · Score: 1

      Perl 6 has failed....

      Rakudo is one of several implementations of Perl 6, and its 28th release is next week. Not only does it exist (and people are using it for real work), but you can also use Perl 5 modules and C shared libraries from it. (Not all of CPAN yet works; that's a work in progress.)

      Oh, and it also does things that Perl 5 does not and probably will not ever be able to do.

  7. Just four years? by Anonymous Coward · · Score: 0

    That's generous.

  8. Erlang is an interesting language by msobkow · · Score: 4, Interesting

    I've been working with Erlang for about 9 months now. It's an interesting language, but prone to some of the most bizarre runtime problems because it doesn't do type checking (for example if you typo a "+" instead of "++" when concatenating strings it'll defer the error to runtime, when it reports an "arith error".)

    One thing that really impresses me about Erlang is how tight the code is. We've been working on a PBX application (with Freeswitch and PostgreSQL) and it's not even 30,000 lines of code in Erlang, including database I/Os and client/server GUI access. C++ would have weighed in at around 100,000 lines for the same functionality.

    --
    I do not fail; I succeed at finding out what does not work.
    1. Re:Erlang is an interesting language by Anonymous Coward · · Score: 2, Interesting

      In many languages ++ means concat the lists, so I assume that if a string is a list of characters then its a concat of the two lists? I'd guess that once you understand the underlying concepts and seen a few functional languages then its not too confusing to debug. It is frustrating to have to wait until runtime to find errors that a decent IDE would identify immediately..

      I always wondered how well Erlang handles caching. An LRU is generally a list crosscutting a hash-table and is done O(1) by reordering the entry. As Erlang uses persistent values, this would be O(n). Also, since an actor handles messages sequentially it couldn't handle a high load. Do Erlang folks do caching in the VM or is it purely externalized? I suspect that the common apporoach is to use memcached with a pool of actors that make the out-of-process calls? This would resolve the shared state issues, though how expensive is the serialization overhead?

    2. Re:Erlang is an interesting language by Algan · · Score: 2, Interesting

      I've been using Erlang as my primary language for the past 2 years and I have to say that you are essentially correct. My worst typo-leading-to-weird-bug situation was forgetting a comma in a list of strings. Apparently ["foo", "bar" "baz"] is interpreted as ["foo", "barbaz"]. That led to some subtle failures in a totally different part of the code.

      However, occasional weird issues aside, it does reduce the size of the code and the number of bugs by a factor of 4 or 5. I have been able to write non-trivial pieces of code in the range of 100 lines that ran fine the first time around. That never happened with C/C++/Java. And the way Erlang and the OTP environment handles product deployments and maintenance is really superb.

      --
      If con is the opposite of pro, is Congress the opposite of progress?
    3. Re:Erlang is an interesting language by Anonymous Coward · · Score: 0

      Don't forget the application frameworks as well.

    4. Re:Erlang is an interesting language by blackfrancis75 · · Score: 1

      uh yeah, I'd happily sacrifice code readability/maintainability/re-use for 'tightness'.
      After all, we run out of room on our source repo all the time, and typing is such a drag!

    5. Re:Erlang is an interesting language by sourcerror · · Score: 1

      Well, there are languages aimed to fix that e.g. Mercury. It's Prolog with types. Although it's been around for 15 years, it's still mainly a research language, and it would be a lot of work to port the OTP environment to there.

    6. Re:Erlang is an interesting language by Anonymous Coward · · Score: 0

      Too bad there's nothing to automatically convert verbose high level languages into tight assembly.

    7. Re:Erlang is an interesting language by gknoy · · Score: 1

      Too bad it's such an ugly language.

      People say the same thing about many other languages that some developers to be remarkably expressive and non-ugly. Lisp and Perl come to mind -- maligned for parentheses or type-indication-prefixes, they each have their strengths. I believe that one of Erlang's strengths is its sparse syntax and ability to express complex ideas in few lines of code. Many would consider that "elegant", even if the syntax seems strange to the uninitiated.

      I've programmed for years in Perl and in a Lisp, and Iquite like them. I've looked enough at an Erlang book to wish I understood it better, but haven't used it for any programming projects. Looking at how it easily handles things which in Java or C++ would require lots of crufty syntax is mindboggling. "That's it? What did I miss?" was a common thought.

    8. Re:Erlang is an interesting language by Procrasti · · Score: 1

      While all assignments in Erlang are constants (or single assignment), the standard pattern is a to have a processing loop that receives and processes messages. This processing loop usually has a single state variable (such as the cache entries in your example), and to alter the state, a new variable is created from the message and the old state, and the loop called (tail recursion stops the stack from growing) with the new state. This means that in practical terms, there is no problem implementing say an O(1) cache in Erlang.

      There is also a mutable, pre process, key/value store (but its use is frowned upon), and also ets (erlang term storage) and dets (disk ets), which are key/value store type systems.

    9. Re:Erlang is an interesting language by Anonymous Coward · · Score: 0

      But how would you perform the LRU reordering in O(1) time? The order is maintained on a doubly-linked list so the adjacent nodes must be mutated. If copied, then it would create a ripple effect so that the entire list is regenerated, thus being O(n) time. I would think an LRU cache would have to be constructed from a different core data structure, such as a persistent tree that has slower access but is cheaper to copy. Or, to give up and encapsulate the mutable variables within the actor's process. These low-level data structure issues are the primary points of confusion that I have with Erlang, as an outsider.

  9. Official site of Perl 6 by Stan+Vassilev · · Score: 1

    Talking about Perl, this is the official Perl 6 site:

    http://perl6.org/

    Srsly :P

  10. 2K38? by the_humeister · · Score: 5, Insightful

    Would it hurt to just write 2038? No space is saved writing it the other way.

    1. Re:2K38? by moteyalpha · · Score: 5, Funny

      I think that 2K-10 is even better. 2K-10=(1024*2)-10=2048-10=2038=(2^11)-10

    2. Re:2K38? by pushing-robot · · Score: 1

      It's a shortened version of "Y2K38", in homage to "the Y2K bug" (which is itself a conveniently shortened version of "the Y2KOMGWTFBBQwereallgonnadie!1!eleven bug").

      --
      How can I believe you when you tell me what I don't want to hear?
    3. Re:2K38? by Trogre · · Score: 1

      That still doesn't stop people saying "double-you double-you double-you" instead of "world wide web", or the dreaded "dub dub dub".

      --
      "Nine times out of ten, starting a fire is not the best way to solve the problem." - my wife
    4. Re:2K38? by Burdell · · Score: 1

      What does "2K-10" mean? How can you subtract 10 from 2 Kelvin? You can't go below absolute zero. Oh, you meant "2k-10", but that's 2000-10 or 1990.

      In any case, using an "abbreviation" that is longer than the original is pretty pointless.

    5. Re:2K38? by Anonymous Coward · · Score: 0

      It's "wub wub wub".

    6. Re:2K38? by QRDeNameland · · Score: 1

      I think that 2K-10 is even better. 2K-10=(1024*2)-10=2048-10=2038=(2^11)-10

      Of course, you should specify that you're using kibiyears.

      --
      Momentarily, the need for the construction of new light will no longer exist.
    7. Re:2K38? by Blakey+Rat · · Score: 1

      Ok, the lameness filter filters tons of genuine jokes, but lets this stinker through? Christ.

    8. Re:2K38? by 3dr · · Score: 1

      I agree. The full octaltechtardic specification for 2038 should be "2KiY-012".

    9. Re:2K38? by moteyalpha · · Score: 1

      No kidding, I didn't even think it was funny when I wrote it. I wish I could mod myself down. Maybe the crowd sourcing has a Y2K+10 bug.

  11. VS2010 & DW by Unka+Willbur · · Score: 1

    Visual Studio 2010 is a very nice step in the right direction, it's much better than previous versions. That said, Dreamweaver CS5 is the first step in a true IDE from Adobe and rocks hard for the huge amount of Wordpress development I do.

    --
    "Remember when I said I would never lie? Well, that was the first time."
  12. CTRL+TAB in Visual Studio 2010 by Anonymous Coward · · Score: 0

    It might be a minor detail, but you should know that they finally fixed CTRL+TAB in VS 2010 so that it behaves like every other tabbed interface in the world.

    ...

    ...

    ...nah, just kidding, it still sucks balls.

  13. Strawberry Perl will be out in a week or so by adamkennedy · · Score: 4, Informative

    Strawberry Perl has been doing betas all the way through the 5.12.0 RC process, so the production release should be out in a week or so.

    What the summary doesn't mention is that there's some stuff in 5.12 that allows Strawberry to add:

    GCC-based 64-bit support for Windows servers

    Strawberry Portable (flash drive) stuff finally works in a first-class manner (with separate core/vendor/site installation targets).

    1. Re:Strawberry Perl will be out in a week or so by Nimey · · Score: 1

      Since I'm too lazy to google (it's almost bedtime), WTF is Strawberry Perl?

      --
      Hail Eris, full of mischief...

      E pluribus sanguinem
    2. Re:Strawberry Perl will be out in a week or so by sowth · · Score: 1

      A strawberry pearl is just a pink one. You see, in some tropical areas, oysters feed on human flesh. All the blood turns the pearls pink.

    3. Re:Strawberry Perl will be out in a week or so by Phroggy · · Score: 3, Informative

      In contrast to ActiveState's ActivePerl (a Windows port of Perl, sponsored by Microsoft, which works well but changes things enough that you generally can't just download some random CPAN module and compile it, you have to use one of the precompiled binary modules they make available), Strawberry Perl is a Windows port of Perl that tries to remain as close as possible to the original UNIX version, but tweaking just enough to get it to work well on the platform. I believe the goal is to move toward Vanilla Perl, which would be essentially taking the plain old normal Perl that runs on UNIX, and just running that on Windows without changing anything.

      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
  14. Definition of 'OTP' by Anonymous Coward · · Score: 0

    First up, reader martinjlogan sends along a tutorial for setting up a workable Erlang/OTP development environment on a Mac. Next, reader acid06 notes news of Perl 5.12, including what may be the first delivered fix for the Y2K38 bug. (Hit the Read More link below for some details on Perl's new release strategy.)

    Here's a helpful link for those wondering what OTP means. I'd never heard it used in this context and I image most other developers haven't either.

    I wish more languages and distributions had out of the box support for OPIE or S/KEY.

    1. Re:Definition of 'OTP' by reillyeon · · Score: 1

      Erlang's OTP stands for "Open Telecom Platform" and is the set of standard libraries and design patterns included with the language. It has nothing to do with authentication.

    2. Re:Definition of 'OTP' by Anonymous Coward · · Score: 0

      Working in electronics I was wondering why the hell would a Mac language be one-time-programmable.

  15. x86 assembler thread by Anonymous Coward · · Score: 0

    In the spirit of merging multiple discussion topics into the same thread, who wants to discuss x86 assembler programming with me? Just hit reply and we'll get under way. Feel free to change the topic at any time.

    1. Re:x86 assembler thread by Yvan256 · · Score: 1

      mov AX, 0013h
      int 10h

    2. Re:x86 assembler thread by QuantumG · · Score: 1

      at&t syntax sucks but at least it is consistent. And ModR/M and SIB bytes are not that complicated, but the Intel manuals sure are painful to read.

      --
      How we know is more important than what we know.
    3. Re:x86 assembler thread by hoggoth · · Score: 1

      You kids with your fancy registers and interrupts. In my day we did:

                ldx #$00
      loop lda message,x
                sta #$0400,x
                inx
                cpx #$07
                bne loop
                rts
      message .text "Welcome"

      --
      - For the complete works of Shakespeare: cat /dev/random (may take some time)
    4. Re:x86 assembler thread by thegrassyknowl · · Score: 1

      X86 assembler programming, isn't the weather outside just splendid today?

      --
      I drink to make other people interesting!
    5. Re:x86 assembler thread by Anonymous Coward · · Score: 0

      mov ax, 0003h
      int 10h

      ; back to textmode!

      mov ax, 2C00h
      int 21h

      ; I think this exited back to dos?
      ; I know COMs could just RET at the end of main(),
      ; but EXEs needed a dos call

  16. What, no ActiveState mention? by Anonymous Coward · · Score: 3, Informative

    While you're waiting for Strawberry Perl to put out a release, why not try a package from a company that has their stuff together? Activestate's ActivePerl 5.12.0 is free-for-non-commercial-use and already out. 32- and 64-bit builds for Windows, Mac, and select Linux distros are available.

    1. Re:What, no ActiveState mention? by Anonymous Coward · · Score: 0

      Last time I looked Actisvestate Perl could not work with CPAN modules. Has that changed now ? Perl without CPAN support is completely useless to me.

    2. Re:What, no ActiveState mention? by Anonymous Coward · · Score: 0

      The goals are different.

      We're just being a little cautious, because we're adding other things to the installer at the same time, and I need to wrap things up and test them before we do a final 5.12 release.

      A beta version should be available at the end of this week.

      (Curtis Jewell - I don't have a login here yet.)

    3. Re:What, no ActiveState mention? by Brownstar · · Score: 1

      Since when?

      Now if you want you can use ActiveStates PPM application, which doesn't contain everything in CPAN) But you can always download the original module from cpan, and make/compile/build it for your version of windows. And it's no more difficult than configuring it for unix/linux barring you have the proper utilities.

      Now you will probably need a Windows compatable make program, such as nmake (ftp://ftp.microsoft.com/Softlib/MSLFILES/nmake15.exe).

      If it's an XS module, or something else that uses C, you might need to have a C compiler, but again tat's no different than if you were running it in unix.

      The only other roadblock, is if the module developer used non-portable coding practises (such as fork) that don't work on Windows. But you're not going to be able to use that module regardless of what Perl build you use on windows.

    4. Re:What, no ActiveState mention? by Anonymous Coward · · Score: 0

      Of course, most of their "getting their stuff together" like, say, releasing within a day or two and having a binary repository that works, and having support for CPAN, has only happened since Strawberry gave them something to compete against :)

  17. VS2010 reinventing personalized menu? by imess · · Score: 1

    Quoting from the Express page:

    Unique to Visual Studio® 2010 Express is a new streamlined user experience that focuses on the most common commands by hiding some of the more advanced menus and toolbars.

    Anyone who has actually downloaded and used it cares to share if this is Windows 2000 style personalized menu all over again?

    1. Re:VS2010 reinventing personalized menu? by trilinear_nz · · Score: 1

      No, they have actually legitimately *removed* certain functionality.

  18. WTONG !! 2002, 2003, 2005, 2008, 2010 by Anonymous Coward · · Score: 0

    Pretend all you want, you can't fool the fool fool!

    Or, 7, 7.1, 8, 9, 10

    I am the one and only, I am the fool fool!

  19. wtf? by MagicM · · Score: 5, Insightful

    Awesome. From now on, let's just post 1 story on Slashdot per day with all of the good stuff in it, so that we may discuss everything in it in one big unrelated clusterfsck of comments.

    1. Re:wtf? by Anonymous Coward · · Score: 0

      For those in Alabama, "one big unrelated clusterfsck" is kind of like the opposite of your family tree.

    2. Re:wtf? by Anonymous Coward · · Score: 0

      You expected something organised and comprehensible when it started with perl?

    3. Re:wtf? by drinkypoo · · Score: 1

      Awesome. From now on, let's just post 1 story on Slashdot per day with all of the good stuff in it, so that we may discuss everything in it in one big unrelated clusterfsck of comments.

      If you have too many real stories on the front page then there's a risk that the slashvertisements will get pushed down and the visitors will never see them. This makes sure that Slashdot's customers get maximum ROI.

      --
      "You're right," Fisheye says. "I should have set it on 'whip' or 'chop.'"
  20. Perl 5.12? by phantomfive · · Score: 1

    I've been waiting nearly a decade for Perl 6. Oh well.

    --
    Qxe4
    1. Re:Perl 5.12? by mr_mischief · · Score: 4, Informative

      Perl 6 is mainly usable, and some form of it is being used in production at multiple sites. It's just not ready for public "launch" yet. If you really want it, you can get it. Perl6.org has it.

      Perl 5 hasn't exactly been sitting still the past decade. The changes between 5.6 and 5.8 or 5.8 and 5.10 are huge. I haven't looked over the full changes list for 5.12 yet, but it sure isn't the language Perl 5 was in 2000.

    2. Re:Perl 5.12? by Anonymous Coward · · Score: 0

      I've been waiting nearly a decade for Perl 6. Oh well.

      Get used to it!
      Perl 5 is the new MacOS 10. ;)

    3. Re:Perl 5.12? by Phroggy · · Score: 1

      I've been waiting nearly a decade for Perl 6.

      Why? Just because you've heard the hype?

      --
      $x='S24;r)>63/* h@<5+oZ)32"5cz';$me='phroggy'x$];
      $x=~y+ -xz+\0-Tx+;print$_^chop$me for split'',$x;
    4. Re:Perl 5.12? by ducomputergeek · · Score: 1

      I heard they were just waiting for the Duke Nuk'em team to finish up. After all, it's Duke Nukem Forever being written in Perl 6 for the Phantom console, right?

      --
      "The problem with socialism is eventually you run out of other people's money" - Thatcher.
    5. Re:Perl 5.12? by Dystopian+Rebel · · Score: 1

      The changes between 5.6 and 5.8 or 5.8 and 5.10 are huge.

      I can only speak for myself, a long-time Perl programmer, but the most significant change for me in this range of versions is Unicode support. The Perl designers added the Switch statement, which I never used, and now in 5.12 they are deprecating it because of problems.

      I find with Perl and C++ that I only use a lucid, practical subset of the possible functionality. I leave all the high-minded theoretical conversations about applying abstruse features to my better educated colleagues. (o;

      I haven't looked over the full changes list for 5.12 yet, but it sure isn't the language Perl 5 was in 2000.

      As a result of my primitive, ignorant approach to programming, I find that Perl in 2010 is very much the same language that I started using 12 years ago. I call this a strength of the language and one of the reasons why learning Perl has been an excellent investment.

      --
      Rich And Stupid is not so bad as Working For Rich And Stupid.
    6. Re:Perl 5.12? by bcrowell · · Score: 1

      Perl 6 is mainly usable, and some form of it is being used in production at multiple sites. It's just not ready for public "launch" yet. If you really want it, you can get it. Perl6.org has it.

      The thing is, perl 5 is a mature language with a mature, high-quality implementation. The last time I ran into a bug in the main implementation of perl 5 was about 8 years ago. Perl 5's implementations have all kinds of crazy features, e.g., you can run it on MacOS 9, and I can make a complete Windows executable of a perl app on my linux box, without even having access to a Windows machine. The collection of Perl 5 modules on CPAN is a mixed bag, but CPAN is mature enough that you can pretty much tell from ratings, comments, and commit histories whether a particular module is high in quality and safe to hitch your wagon to.

      All of this quality makes me stay with perl, even though python and ruby have prettier syntax.

      Perl 6 has many years to go before it reaches this level of quality. First they have to get a feature-complete, debugged release out, and update the O'Reilly book so it describes the language in the actual form it ends up in. Then the whole community has to go through all the years of work required to bring the perl 5 implementation up to the same high standard of quality as perl 5. The thing is, by the time that happens, python and ruby will have matured a lot more. I'm not even sure I would switch from perl 5 to perl 6. I might switch from perl 5 to ruby. I like ruby, but my experience so far is that it just isn't mature enough.

      Perl 6 was also originally supposed to give us a pseudocode that would be shared with other languages. That was, for me, the main reason to be excited about it 10 years ago. The fact that progress has been so slow has meant that the other language communities have given up on cooperating on that idea, so it's not going to happen. That makes me a lot less excited about perl 6 than I was back when I first heard about it -- which was when my now-teenaged daughter was three years old.

    7. Re:Perl 5.12? by chromatic · · Score: 2, Interesting

      Perl 6 was also originally supposed to give us a pseudocode that would be shared with other languages.

      Parrot? It's out. It works.

    8. Re:Perl 5.12? by iggymanz · · Score: 1

      fanboys and shills....

      "mainly usable" "some form of it"

      quit kidding yourself, it's not done yet, still being designed, is changing, and no one with a brain will use a language for production for which there are many forms because *it isn't done yet*.

  21. VS2010 bug by BradleyUffner · · Score: 1

    Visual Studio 2010 still hasn't fixed one of the major bugs that has been around since .NET 1.0.
    When you run a forms based program in debug mode on an x64 system, and the form's load event throws an exception, the program will happily continue running without reporting the exception. Execution skips directly to the end of the load sub without running any lines of code after the exception, yet the program continues to run as if nothing went wrong. Everything works as expected if you are working on an x86 box though. You will rip your hair out trying to figure out what's going on if you don't know about this.

    Details can be found here http://social.msdn.microsoft.com/Forums/en-SG/vsdebug/thread/69a0b831-7782-4bd9-b910-25c85f18bceb

    1. Re:VS2010 bug by Anonymous Coward · · Score: 1, Informative

      Actually, the post you mention points out that this is a bug in x64 Windows, not in Visual Studio.

  22. Re: new major version of Perl is now available. by Sarten-X · · Score: 4, Insightful

    It's considered trolling because that's purely what it is.

    Perhaps you'd like to enlighten us with some examples of "byzantine syntax", since Greek and Latin aren't valid Perl. Perhaps you meant "British" syntax, since the language is based more off of English than other programming languages. Yes, I'm aware this places it in the same group as COBOL, but hear me out.

    Perl's syntax is based off natural language. You should be able to tell the computer what to do in the same way you'd tell another person, with some obvious extra clarity needed. In this regard, it is the opposite of INTERCAL. In Perl, when you want to run a function 'foo' if and only if 'bar' is true, the statement is very close to what I've just said.

    foo if bar;

    Sure, it's not like C, but why should we limit our thinking? Alternatively, just as in English you can say "If 'bar' is true, run 'foo'":

    if (bar) { foo }

    Perl is a language for people who use language. Even the more ugly syntax makes perfect sense with a bit of thought:

    s/foo/bar/g;

    That's a full statement, saying "substitute foo with bar globally". It applies to whatever you happen to be working on at the moment, just like how there are many statements in English that carry an implied 'you'.

    The syntax of the Perl language is only ugly if you try to forget that it is a language. Perhaps other languages should try to emulate Perl's features, and actually gain some readability. Sometimes, it just makes more sense to express things differently. Would COBOL be so bad if it allowed BASIC's syntax as well (and allowing non-computer people to write in it carried the death penalty)?

    --
    You do not have a moral or legal right to do absolutely anything you want.
  23. Re: new major version of Perl is now available. by QuietLagoon · · Score: 1
    byzantine -- (sometimes l.c.) characterized by elaborate scheming and intrigue

    Describes it rather well for me.

  24. Re: new major version of Perl is now available. by Sarten-X · · Score: 1

    If Perl's syntax were actually elaborate, that might hold. Maybe something like "every third line MUST be blank, unless it's a full moon, or a Thursday, in which case being blank is optional, but the line must include an uppercase letter."

    Offering more than one way to write a given line, and attempting to follow the rules of a natural language, is not "elaborate scheming and intrigue". Perhaps you have a counter-example? or is this just more FUD from the I-was-told-Perl-is-ugly-so-I'll-repeat-it school of thought?

    --
    You do not have a moral or legal right to do absolutely anything you want.
  25. Just to clarify... by Valdrax · · Score: 5, Funny

    %^$%^$%^$%^$%^ NO CARRIER

    Just to clarify, does that actually do something in Perl?

    --
    If it's for-profit but free, you're not the customer -- you're the product (e.g., the Slashdot Beta's "audience").
    1. Re:Just to clarify... by rthille · · Score: 1

      And my freaking mod points expired...

      --
      Awesome furniture, accessories and cabinetry in Santa Rosa, CA: http://humanity-home.com/
    2. Re:Just to clarify... by gknoy · · Score: 1

      Sadly, it gives a compiler error:

      ~$ perl -e '%^$%^$%^$%^$%^ NO CARRIER'
      Scalar found where operator expected at -e line 1, near "%^$%"
                      (Missing operator before $%?)
      syntax error at -e line 1, near "%^$%"

    3. Re:Just to clarify... by Anonymous Coward · · Score: 0

      So, is that an error message or was perl cursing at you?

  26. IntelliSense Overhall by ponraul · · Score: 1

    Is this finally going to be the version where Intellisense doesn't make the editor randomly pause and unresponsive?

    I usually end up disabling Intellisense in C++ by removing the offending DLL from the VS installation when I'm working with heavily <template>ed code.

    1. Re:IntelliSense Overhall by serviscope_minor · · Score: 1

      Good luck with that. Unless intellisense gets fast enought to solve the halting problem, you can write template code that it will take literally forever to parse.

      --
      SJW n. One who posts facts.
  27. no brainer? by thoughtsatthemoment · · Score: 4, Informative

    I've been using VC++ Express 2008 and was excited to get an upgrade. But instead I was surprised by a few seemingly "dumb" moves:

    1. The default fonts for the editor and tool windows have been changed to a font that looks very blurry on Windows XP. To change them back, you have to change them one by one for every window.

    2. The drag'n Drop capability to add buttons to the tool bars is gone. You have to find the button from another dialog and then click "MoveUp/Down" several times to move it to the place you want.

    3. The GUI I used the most in the Option Dialog, the directories of Exe/Include/Lib, is moved to a place that I haven't yet found.

    4. The startup time is much longer than that of 2008.

    5. The new GUI has a high contrast. Maybe it's just me, but after staring at it for a long time, I feel like I am starting to see ghost images.

    1. Re:no brainer? by Anonymous Coward · · Score: 0

      The exe/include/lib stuff is now per-project only, I believe. Yes, it's insane.

    2. Re:no brainer? by yuriks · · Score: 1

      To change those settings you must now do this (confusing, yeah):

      1. Open the Property Manager (View -> Property Manager)
      2. Expand your project and any of the setting groups
      3. Edit Microsoft.Cpp.Win32.user (VC++ Directories)

      Yeah, it's quite confusing, but that setting is global.

    3. Re:no brainer? by heffrey · · Score: 1

      Your blurry fonts are likely an issue with Cleartype. Are you using a CRT monitor?!

      Also, what kind of loser still runs XP in 2010?!

    4. Re:no brainer? by thoughtsatthemoment · · Score: 1

      I don't know what kind they are, but there are a lot of them: http://arstechnica.com/microsoft/news/2010/04/windows-7-surpasses-10-market-share.ars

      And are you one of those mediocre programmers who are just lucky enough to be developing cleartype but fail to realize the best way to promote a technology is to improve it instead of relentlessly push it?

    5. Re:no brainer? by Anonymous Coward · · Score: 0

      4. The startup time is much longer than that of 2008.

      Posting AC since I moderated.

      For me, VS 2010 is an order of magnitude faster than VS2008. I had a solution file that was eating 700MB of memory in VS2008 after initial load. Right now, that same solution is using 78MB in VS2010. Sometimes it shoots up to 400MB then goes back down.

      It also loads the solution faster. This is on Windows XP 32-bit.

    6. Re:no brainer? by thoughtsatthemoment · · Score: 1

      By start up time I meant to launch VS without any solution. I believe it has something to do with the managed code used in 2010 that requires the extra time. I maybe a little picky here since on a fast machine like an extreme version of xeon, it's only like 1 or 2 seconds longer (I tested it first with an ordinary dual core)

    7. Re:no brainer? by rdnetto · · Score: 1

      4. The startup time is much longer than that of 2008.

      You can thank WPF for that. I took a look at it when it first came out, but even the most trivial app takes 1+ sec to start up, whereas WinForms apps start virtually instantaneously.
      Amusingly, Qt has had the XML-based GUI representation that Microsoft tried to achieve with WPF for way longer, and it's also reasonably fast.

      --
      Most human behaviour can be explained in terms of identity.
  28. Yeah it's a problem - CE6 by Weaselmancer · · Score: 1

    CE6 development is no longer done through Platform Builder. It's a Visual Studio plugin. But - only for VS 2005. They've never updated it. Even if I wanted to upgrade I couldn't.

    --
    Weaselmancer
    rediculous.
  29. Re: new major version of Perl is now available. by Anonymous Coward · · Score: 1, Insightful

    Things that suck: the implicit variables and default pattern space, the special variables with a sigil and another punctuation character, the fact that there are sigils and subroutines without named parameters, the irritating hassle that is hashes of hashes, the ugly shitball bolted on OO support. Nor is CPAN is a plus, with the thousands of modules that partially and/or totally duplicate one another's functionality or that fail to work owing to years of neglect or, more often, carelessness of packaging on the programmer's part.

    But the worst part about perl is the ethic of TMTOWTDI. In practice means either that one finds it done a new way for every 1000 or so lines, or that someone's imposed exactly their way to do it. Your programming is not art, and you are not writing literature; you are not a code poet. Expressing yourself in code is nowhere nearly as important as writing something stable and maintainable. One - just one - good, obvious way to do it; something that's predictable and that works. Something that, when you've moved along, the next guy can look at and not have to ask aloud what fucking asshole did this.

    There's better choices out there now, even among scripting languages. We don't need to keep using the monstrous bastard amalgam of sh, sed, awk and Larry Wall's pretenses of being a linguist. Learn some new skills, and let's all just let the fucking thing die already.

  30. Is VS2010 still slow? by MrCrassic · · Score: 3, Informative

    A few days ago, I used a copy of Visual Studio 2010 that I got from my MSDN Academic Alliance account. It looks really nice, but it ran absolutely dog slow. And this was for debugging VBScript!

    I gave it another shot with Visual Studio 2010 Web Developer Express, which I heard can debug VBScript just like the full devenv can. It was a little faster (though still slower than VS2008), but it nor Visual Basic 2010 Express would debug my VBScript.

    I haven't tried coding on it for real (I also do C/C++ development; can't wait to port that script over to a REAL effin' language), but if it's as slow as I remember it being, I can see lots of companies turning back really quickly.

    1. Re:Is VS2010 still slow? by TrancePhreak · · Score: 1

      For a developer, your question has very few details. Kind of like asking why does my program crash without any details.

      Anyways, post some details.

      --

      -]Phreak Out[-
    2. Re:Is VS2010 still slow? by julesh · · Score: 1

      I used a copy of Visual Studio 2010 that I got from my MSDN Academic Alliance account. It looks really nice, but it ran absolutely dog slow.

      Great. And VS2005 slows my machine to a crawl if you use it wrong. Have a ~2000 line C# source file in your project, and intellisense can take about 5-10 seconds to open. Makes you regret asking for it... would've almost been faster to look the name up in the documentation and type it. Really looking forward to the day I'm forced to upgrade to an even slower version.

    3. Re:Is VS2010 still slow? by MrCrassic · · Score: 1

      My script is less than 2000 lines, but it somehow took 10+ seconds to open. Typing would be incredibly sluggish too. Doing anything in the GUI was a mistake. This is on a P8600 Core 2 Duo with 4GB of RAM.

  31. Bugs? Impossible! by AliasMarlowe · · Score: 5, Funny

    bugs would creep in, undoubtedly

    Impossible, if you follow Micosoft's guidelines, such as those published in the Microsoft Programming Series book by Steve Maguire: "Writing Solid Code: Microsoft Techniques for Developing Bug-free C" ISBN 978-1556155512
    Microsoft uses these very techniques themselves, in every single one of their bug-free programs. Uh, on second thoughts...

    --
    Those who can make you believe absurdities can make you commit atrocities. - Voltaire
  32. Re: new major version of Perl is now available. by ais523 · · Score: 2, Informative

    I'm offended by your statement that it's impossible to write valid Perl in Latin, and offer this counterexample. Arguably, a Latinised syntax fits it even better than an English-like one; in both Perl and Latin, there's often enough context to put things in a (relatively) arbitrary order. Of course, I'm someone whose signature is an INTERCAL statement, so you may want to take what I've said with that in mind.

    --
    (1)DOCOMEFROM!2~.2'~#1WHILE:1<-"'?.1$.2'~'"':1/.1$.2'~#0"$#65535'"$"'"'&.1$.2'~'#0$#65535'"$#0'~#32767$#1"
  33. Re: new major version of Perl is now available. by plasticsquirrel · · Score: 0, Flamebait

    Alright, I'll bite. Perl is so byzantine that its creators can barely write a compiler for the next version of it. Perl is so byzantine that its syntax can be meaningfully compared with the sloppy mess called the English language. Perl is so byzantine that most of the people using Perl left it for Python a decade ago. Perl is so byzantine that its syntax is almost half as bad as Haskell's. ;-)

    It seems like Perl is a byproduct of another era, when being "quicker to write than C" was a big feature. Since anyone can just use Python and read an easy "spam and eggs" tutorial, and do most of the same things with cleaner syntax, what is the big draw? The Lisps had all the advanced features long ago, and the other friendlier "scripting languages" are here to stay. What's the future of Perl? Can it really survive as a popular hacker language like it was 10-15 years ago?

    --
    Systemd: the PulseAudio of init systems
  34. Re: new major version of Perl is now available. by kwoff · · Score: 1

    Different people think differently. Clean syntax and propriety might be important to you. For others, it might feel less expressive, like trying to speak in a foreign language.

  35. Erlang for Macintosh by gnasher719 · · Score: 1

    Just wondering... Is it so difficult to put the whole thing into a package that will install with a double click?

    So we have a tutorial to download the sources and install it. Which uses a package manager which is "quickly becoming the package manager of choice for OSX" except that I have never heard of it. Which comes with another link for download instructions. Look, we don't want links to download instructions, we want a link that downloads the whole package which installs with a double click.

    Someone has to learn a bit about user interfaces.

  36. Re: new major version of Perl is now available. by Anonymous Coward · · Score: 0

    The sigils can be argued to be a boon that makes code clearer (this is a preference (see (e.g.) lisp)).

    For OO, check Perl 5's Moose (more or less Perl 6 OO system backported). It arguably beats the pants of Python and Ruby.

    See MooseX::Declare for a nice syntax. Including parameter declarations for modules.

    CPAN can be a pain, but it is still the gold standard for what it does.

    There is more than one way.. Perl is rich; that is a design choice. The main disadvantage is learning time. The main advantages are expressability and that it is fun.

    And so on. It isn't worth the time; too many trolls. Even if you got the facts more or less correct.

  37. ... for methods by Anonymous Coward · · Score: 0


    s/for modules./for methods./;

    Sorry about that.

  38. Re: new major version of Perl is now available. by Anonymous Coward · · Score: 0

    Perhaps you'd like to enlighten us with some examples of "byzantine syntax", since Greek and Latin [wikipedia.org] aren't valid Perl.

    Wanna bet?

  39. Re: new major version of Perl is now available. by jonadab · · Score: 2, Insightful

    > since Greek and Latin aren't valid Perl

    Well, there is Lingua::Romana::Perligata. Predictably, we have Damian to blame for that.

    --
    Cut that out, or I will ship you to Norilsk in a box.
  40. VS2010, a fresh new look... by gabereiser · · Score: 0

    VS2010 is really a whole revamp of the traditional visual studio. I'm been using it now for a few months and I must say, once you get used to the new layout and views... ...it's far better than 2005 or 2008. There are (of course, with microsoft) a few crashing bugs dealing with intellisense but still. Very easy to use and makes other IDE's look legacy now. If only XCode had intellisense the way Visual Studio does. (I know about hitting escape within XCode, still, it's not the same).

  41. Re: new major version of Perl is now available. by Anonymous Coward · · Score: 0

    Perl's syntax is based off natural language. You should be able to tell the computer what to do in the same way you'd tell another person, with some obvious extra clarity needed.

    An unparsable language is not exactly what I would think of first, when "extra clarity" is required.

  42. Re: new major version of Perl is now available. by Anonymous Coward · · Score: 0

    It's considered trolling because that's purely what it is.

    Perhaps you'd like to enlighten us with some examples of "byzantine syntax", since Greek and Latin aren't valid Perl.

    Lingua::Romana::Perligata. You were saying?

  43. Re:Python is better by Anonymous Coward · · Score: 0

    Most of GUI utilities for programming python are able to scan the current namespace and do autocompletion based on that information. This is retrieved from python runtime.

    IMHO this is much better approach than propaganda intellisense crap.

  44. Given how Prolog-like Erlang is... by weston · · Score: 1

    ... and its at least soft real-time support, we actually got pretty close here.

    1. Re:Given how Prolog-like Erlang is... by Haxamanish · · Score: 1

      Thank you, your comment made me read up on Erlang, and you're absolutely right. I'll give Erlang a try. I was disappointed too soon, slashdot remains a beacon of relevance.

  45. Re: new major version of Perl is now available. by Sarten-X · · Score: 1

    I stand corrected. I should have checked CPAN. Of course there's more than one language to do it in.

    --
    You do not have a moral or legal right to do absolutely anything you want.
  46. Re: new major version of Perl is now available. by Sarten-X · · Score: 1

    Try reading your own link. Perl, a dynamic language, cannot be parsed statically. In other news, green is not red, and water is wet.

    --
    You do not have a moral or legal right to do absolutely anything you want.
  47. Not being a serious Microsoft Developer is good by Douglas+Goodall · · Score: 1

    For several decades I bought everything Microsoft. Each upgrade I could hardly wait for the new stuff. After a while it became clear there was no end, and as the window of opportunity for each version got shorter and shorter, it became clear that a single developer or team could not keep up Microsoft's relentless updates. It didn't even seem like several teams alternating release cycles could keep up. If every release takes a six month learning curve, and something changes every five months of so, there is no end, or even a plateau on which to breath. To make things worse, a collection of technical materials and training are required to get critical mass on Microsoft development at any particular moment. At a certain point I released I could not read the books fast enough to keep up, and everywhere I looked in my lab were piles of Microsoft press books at 60-80 dollars apiece, sitting around like paperweights. The books have become hopelessly version oriented, and become obsolete in just a few months. anyway, the point I am trying to make is that to be a "serious Microsoft shop", takes a serious amount of money and staff, and under the circumstances, I understand why companies do not want to afford in-house programming staff. Microsoft has their hand in the deep pockets of american business, and it is a major drag, lie taxes, on the economy and the lifeblood of businesses. It is one thing, if a company sells tools that are powerful and efficient, but in my opinion, Microsoft doesn't care about the welfare of developers, and only seem to care about the next credit card. I saw the earth shift on it's axis when I realized Microsoft saw developers as another class of paying customers, and not the strategic partners developers once were. Bah humbug, not another penny for Microsoft, that's my motto.

  48. Re: new major version of Perl is now available. by chromatic · · Score: 1

    Perl is so byzantine that its creators can barely write a compiler for the next version of it.

    In truth, there are several implementations of Perl 6.

  49. my bank runs Perl 6 on HURD on a CRAY-4 by rubycodez · · Score: 1

    you name of one of many partial implementations of Perl 6, which isn't done yet. To repeat, no sane person nor competent business is going to use Perl 6 or any other half-baked science project for real business.

    1. Re:my bank runs Perl 6 on HURD on a CRAY-4 by chromatic · · Score: 1

      "Done" is an arbitrary, arbitrary word, like "production" or "stable". It means only and exactly what you want it to mean. Perl 5 isn't done yet either, nor is C99 (at least if you use Visual Studio).

      Meanwhile, Rakudo Perl 6 passes more tests today than it did yesterday, and it runs faster than it did yesterday. You can do more with next week's release than you could with last month's release. None of this is accidental nor surprising.

    2. Re:my bank runs Perl 6 on HURD on a CRAY-4 by rubycodez · · Score: 1

      The Perl team itself has definitions of "production" and "stable" for their wares, they have test suite and regression tests, not arbitrary at all whether a build of the language passes or fails. versions such as 5.8.8 are used by commercial and private sector companies, even including banks and financial institutions.

      Someday Perl 6 might get there, though it's becoming clear that project is a failure due to lack of focus and all-inclusive feature creep by Larry

    3. Re:my bank runs Perl 6 on HURD on a CRAY-4 by chromatic · · Score: 1

      ... not arbitrary at all whether a build of the language passes or fails.

      The monthly Rakudo releases pass their tests. Occasionally a couple of tests fail for a day or two during the month. Readers are welcome to make up their own minds about the efficacy of this particular definition of "stable" and "production". (Don't use the HEAD revision in production unless you're willing and able to diagnose bugs.)