Exchange 2013



  • So IT is having an issue with our Exchange environment. Outlook cannot connect to Exchange to retrieve new mail. I've seen Outlook Office and our Outlook Web edition react differently under the same circumstances, so I thought to give the web edition a try. The result:

    <FONT color=blue size=15>:-(</FONT>

    Yes, apparently the new Microsoft standard for the 'Error' red cross, is a sad smiley... Also, the error message on this page is very helpful indeed:

    "something went wrong" (Capitalisation and punctuation [sic])
     
    I feel dumber just having to look at this shit.

  • FoxDev

    @steenbergh said:

    Yes, apparently the new Microsoft standard for the 'Error' red cross, is a sad smiley... Also, the error message on this page is very helpful indeed:

    "something went wrong" (Capitalisation and punctuation [sic])
     
    I feel dumber just having to look at this shit.
    Imagine you're a marketing drone at Shiny Products Corp. What would you rather see in an error message: a simple statement something went wrong, or a wall of technical nonsense that could mean anything from summoning Satan to a delicious cake recipe?
     
    Also, you haven't mentioned if there's a link to get more info about the error.

     



  • Hah, a link to get more information? Of course not. There is, however, a link that says 'Refresh the page now.' I guess for people who can't find the refresh nutton in their browsers or are unaware of F5?...



  • @steenbergh said:

    Hah, a link to get more information? Of course not. There is, however, a link that says 'Refresh the page now.' I guess for people who can't find the refresh nutton in their browsers or are unaware of F5?...

    I'm not quite sure what you expect to see from the client software. It's not exactly rare for server-client-software to suppress all error messages when something goes wrong on the server.

    Alternatively, they could've served up a simple 503 status page which essentially amounts to the same: "Something went wrong."


  • FoxDev

    @steenbergh said:

    refresh nutton
     

    I'd love to hear more about this refresh nutton :P

     



  • The march of catering to the LCD continues. Move along.


  • Trolleybus Mechanic

    @RaceProUK said:

    @steenbergh said:

    refresh nutton
     

    I'd love to hear more about this refresh nutton :P

    I fucking love those, especially with a side of korma and some hot naan. 

     



  • @Rhywden said:

    @steenbergh said:

    Hah, a link to get more information? Of course not. There is, however, a link that says 'Refresh the page now.' I guess for people who can't find the refresh nutton in their browsers or are unaware of F5?...

    I'm not quite sure what you expect to see from the client software. It's not exactly rare for server-client-software to suppress all error messages when something goes wrong on the server.

    Alternatively, they could've served up a simple 503 status page which essentially amounts to the same: "Something went wrong."

    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users? Because the average office drone has the computer skills of stale cheese? Isn't it time that the masses embrace their ignorance when something unexpected appears on the screen and notify someone with more than two braincells?

    Further; someting can go wrong, sure. And from a security perspective it might have value to not display everything in detail. But a bloody SMILEY? I'm at the office, not on Facebook... It's insulting and unprofessional.



  • @RaceProUK said:

    @steenbergh said:

    refresh nutton
     

    I'd love to hear more about this refresh nutton :P

     

    OK, a 'nutton' mght be harder to find than I implied. But I meant to type 'button'. Guess that one slipped through CS's spell checker...


  • Discourse touched me in a no-no place

    @steenbergh said:

    Guess that one slipped through CS's spell checker...
    CS's what?



  • @steenbergh said:

    So IT is having an issue with our Exchange environment. Outlook cannot connect to Exchange to retrieve new mail. I've seen Outlook Office and our Outlook Web edition react differently under the same circumstances, so I thought to give the web edition a try. The result:

    <font color="blue" size="15">:-(</font>

    Yes, apparently the new Microsoft standard for the 'Error' red cross, is a sad smiley... Also, the error message on this page is very helpful indeed:

    "something went wrong" (Capitalisation and punctuation [sic])
     
    I feel dumber just having to look at this shit.
    I hate to sound like yellow Lotus guy, but that's only a default. It's a sensible default, too. It may annoy you not to know what's going on, but what are you going to do with the knowledge if you had it? If you have access to the servers to fix things, you have access to the proper error logs.

    The amount of support time that this saves dealing with ordinary users must be immense.


  • FoxDev

    @steenbergh said:

    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users?

    Because end users won't know how to deal with 'Server frob wasn't binged before first smenting'. Or something like that.


  • Considered Harmful

    @RaceProUK said:

    @steenbergh said:

    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users?

    Because end users won't know how to deal with 'Server frob wasn't binged before first smenting'. Or something like that.


    Sure they will. Ask Ted, or, if slightly more ept, Google it.



  • @steenbergh said:

    @Rhywden said:

    @steenbergh said:

    Hah, a link to get more information? Of course not. There is, however, a link that says 'Refresh the page now.' I guess for people who can't find the refresh nutton in their browsers or are unaware of F5?...

    I'm not quite sure what you expect to see from the client software. It's not exactly rare for server-client-software to suppress all error messages when something goes wrong on the server.

    Alternatively, they could've served up a simple 503 status page which essentially amounts to the same: "Something went wrong."

    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users? Because the average office drone has the computer skills of stale cheese? Isn't it time that the masses embrace their ignorance when something unexpected appears on the screen and notify someone with more than two braincells?

    Further; someting can go wrong, sure. And from a security perspective it might have value to not display everything in detail. But a bloody SMILEY? I'm at the office, not on Facebook... It's insulting and unprofessional.

    Because the average end user is not actually able to use the descriptive error message in any useful way. And if you're able to fix things, then you have access to the descriptive logs anyway.

    Not to mention that, as you yourself pointed out, it's not the best idea to yield full-blown error messages to a potentially public terminal due to security reasons.



  • @Lorne Kates said:

    @RaceProUK said:

    @steenbergh said:

    refresh nutton
     

    I'd love to hear more about this refresh nutton :P

    I fucking love those, especially with a side of korma and some hot naan.
     

    Divide 1 / 0 and stick the result in the microwave for a minute or two, and you'll have hot NAN.

     



  • @steenbergh said:

    But a bloody SMILEY? I'm at the office, not on Facebook... It's insulting and unprofessional.
    I guess you haven't seen the Windows 8 BSOD yet then.



  • @ender said:

    I guess you haven't seen the Windows 8 BSOD yet then.
     

    To be honest, I've been using Win8.0 for about a year, and I haven't seen it.


  • FoxDev

    @dhromed said:

    @ender said:

    I guess you haven't seen the Windows 8 BSOD yet then.
     

    To be honest, I've been using Win8.0 for about a year, and I haven't seen it.

    I have, but only when trying to set up one of the preview editions on Virtual PC (on Win7), which doesn't support Win8.

     



  • @steenbergh said:

    Further; someting can go wrong, sure. And from a security perspective it might have value to not display everything in detail. But a bloody SMILEY? I'm at the office, not on Facebook... It's insulting and unprofessional.

    Then maybe Chrome should determine if the browser is used at the office or on Facebook before showing this:


  • Trolleybus Mechanic

    [url="http://thedailywtf.com/Comments/I-Am-Error.aspx#420361"]CIRCULAR DEPENDENCY![/url]

     



  • @RaceProUK said:

    @steenbergh said:
    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users?

    Because end users won't know how to deal with 'Server frob wasn't binged before first smenting'. Or something like that.

    True. But then there is the case of programs which display a "descriptive" error messages that is wrong.  This just happened with the latest version of Thunderbird. When attempting to access the configuration settings you may get an error message which says: "<font face="courier new,courier">The path C:\Users\username\AppData\Roaming\Thunderbird\Profiles\profile.name\Mail\Local Folders is not suitable for message storage</font>" and you cannot edit or even view any of the configuration settings. The problem with this error message is that the path is correct. That's the standard location that Thunderbird uses and has always used (on Windows).

     So even if you are a fairly knowledgeable person, you still have to do some Goggling to figure out what the problem is because the "descriptive" error message makes no sense.



  • @Lorne Kates said:

    CIRCULAR DEPENDENCY!

     

    YOU'RE MAKING ME DIZZY!!
        \

     


  • Considered Harmful

    @El_Heffe said:

    @RaceProUK said:

    @steenbergh said:
    I'm getting sick and tired of being kept away from my descriptive error messages. Why should this be held from end users?

    Because end users won't know how to deal with 'Server frob wasn't binged before first smenting'. Or something like that.

    True. But then there is the case of programs which display a "descriptive" error messages that is wrong.  This just happened with the latest version of Thunderbird. When attempting to access the configuration settings you may get an error message which says: "<font face="courier new,courier">The path C:\Users\username\AppData\Roaming\Thunderbird\Profiles\profile.name\Mail\Local Folders is not suitable for message storage</font>" and you cannot edit or even view any of the configuration settings. The problem with this error message is that the path is correct. That's the standard location that Thunderbird uses and has always used (on Windows).

     So even if you are a fairly knowledgeable person, you still have to do some Goggling to figure out what the problem is because the "descriptive" error message makes no sense.


    Kind of like the Community Server error message that assumes the problem must lie within SiteUrls.config (or some such)?


  • Trolleybus Mechanic

    @joe.edwards said:

    Kind of like the Community Server error message that assumes the problem must lie within SiteUrls.config (or some such)?
     

    What if there IS a problem with SiteUrls.config for real? I bet you'll feel like a real assho then.



  • @Lorne Kates said:

    assho

    Usually they charge extra for that.


  • Discourse touched me in a no-no place

    @Mason Wheeler said:

    Divide 1 / 0 and stick the result in the microwave for a minute or two, and you'll have hot NAN.
    I'm sorry, but just gets you (positive) Inf. For NaN, you need to divide zero by zero.



  • @dkf said:

    @Mason Wheeler said:
    Divide 1 / 0 and stick the result in the microwave for a minute or two, and you'll have hot NAN.
    I'm sorry, but just gets you (positive) Inf. For NaN, you need to divide zero by zero.
    That depends. You might have negative zero.


  • Discourse touched me in a no-no place

    @Ben L. said:

    That depends. You might have negative zero.
    It doesn't matter when it comes to NaN. (Except on certain HP hardware, IIRC. “Thanks, HP!” But I've never used any of those CPUs; just dealt with support requests relating to them.) Negative zero comes from preserving the sign bit in operations that round toward zero (from below, of course) due to significance loss.



  • @dkf said:

    @Ben L. said:
    That depends. You might have negative zero.
    It doesn't matter when it comes to NaN. (Except on certain HP hardware, IIRC. “Thanks, HP!” But I've never used any of those CPUs; just dealt with support requests relating to them.) Negative zero comes from preserving the sign bit in operations that round toward zero (from below, of course) due to significance loss.


  • Discourse touched me in a no-no place

    @Ben L. said:

    You make my point for me. (How do you think floating-point division is actually performed in hardware?)



  • @dkf said:

    @Ben L. said:
    @dkf said:
    @Ben L. said:
    @dkf said:
    @Mason Wheeler said:
    Divide 1 / 0 and stick the result in the microwave for a minute or two, and you'll have hot NAN.
    I'm sorry, but just gets you (positive) Inf. For NaN, you need to divide zero by zero.
    That depends. You might have negative zero.
    It doesn't matter when it comes to NaN. (Except on certain HP hardware, IIRC. “Thanks, HP!” But I've never used any of those CPUs; just dealt with support requests relating to them.) Negative zero comes from preserving the sign bit in operations that round toward zero (from below, of course) due to significance loss.

    You make my point for me. (How do you think floating-point division is actually performed in hardware?)

    Okay, let's step through this one carefully.

    Mason: Divide 1 by 0 and stick it in your microwave and you get hot NAN.
    dkf: No, you get +Infinity
    Ben: You can also get -Infinity
    dkf: No, you get NaN
    Ben: [picture of getting +Infinity and -Infinity]
    dkf: Exactly. That proves the point I made in my previous post.
    Ben: Okay, let's step through this one carefully.



  • @steenbergh said:

    So IT is having an issue with our Exchange environment. Outlook cannot connect to Exchange to retrieve new mail. I've seen Outlook Office and our Outlook Web edition react differently under the same circumstances, so I thought to give the web edition a try. The result:

    <font color="blue" size="15">:-(</font>

    Yes, apparently the new Microsoft standard for the 'Error' red cross, is a sad smiley... Also, the error message on this page is very helpful indeed:

    "something went wrong" (Capitalisation and punctuation [sic])
     
    I feel dumber just having to look at this shit.

    They're just following the standards set by the Windows 8 BSOD:

    Now, let me rant about it for a while, because I'm in a rant-about-new-MS-politics mood today:

    1. A sad smiley is generally a stupid idea for a system message. What's next, "d00d ur c0mp00ter has bn pwn3d" error?

    2. We're just collecting some error info, and then we'll restart for you? How sweet of them! What a nice, calm, soothing message! That must mean this "problem" they talk about is not a big deal! Wrong. There's a reason Blue Screen of Death is called "Of Death": it should be fucking scary. It should make the end-user piss their pants (figuratively) and call for help (literally), because there's something seriously wrong. As it is now, the user is likely to dismiss the BSOD if it doesn't keep happening too often, but a randomly-appearing BSOD usually is a symptom of some underlying problem that can only grow bigger (faulty RAM stick, overheating, you name it).

    3. My "favourite" part: If you'd like to know more, you can search later online for this error:. First of all, it's utterly uninformative: up to this point, BSOD at least showed you the name of the faulty driver/process/etc, and some additional parameters narrowing the diagnosis a bit. Second: it treats the user like a moron. Not all Windows 8 users are computer illiterates who can only GooBing for stuff, to some of them, the message might actually be of some meaning. If I were one of them (and sometimes I am), I'd be pretty offended by Microsoft telling me to Bing off.



  • @Ben L. said:

    Mason: Divide 1 by 0 and stick it in your microwave and you get hot NAN.


    dkf: No, you get +Infinity


    Ben: You can also get -Infinity


    dkf: No, you get NaN


    Ben: [picture of getting +Infinity and -Infinity]


    dkf: Exactly. That proves the point I made in my previous post.


    Ben: Okay, let's step through this one carefully.

    What he's saying is that your -0 is actually implemented (according to IEEE 754, may it writhe in hell) with a sign bit that can be negative even if the number is 0. As such, -0 can be represented. Now, look at the variable decomposition of the second JS snippet:


    t1 = 1

    t2 = -0

    t1/t2

    Assuming that variable assignment happens immediately, in order for this to be -infty, t2 must retain the sign bit, or else you'd just get infinity. This is the point that dkf is trying to make.



  • @Mason Wheeler said:

    Divide 1 / 0 and stick the result in the microwave for a minute or two, and you'll have hot NAN.
    When I tried that, it only made my microwave implode.

    You bastard.

    Anyway, does anyone know a good way to get rid of a micro black hole in your kitchen?



  • @Anonymouse said:

    Anyway, does anyone know a good way to get rid of a micro black hole in your kitchen?
     

    Try not to throw things into its event horizon, and allow it to evaporate on its own.



  • @lushr said:

    Assuming that variable assignment happens immediately, in order for this to be -infty, t2 must retain the sign bit, or else you'd just get infinity. This is the point that dkf is trying to make.
    Hmm, but 0 === -0...

    17:39:30.598var t1 = 0, t2 = -0;
    17:39:30.601undefined
    17:39:39.8231 / t1
    17:39:39.827Infinity
    17:39:42.3711 / t2
    17:39:42.375-Infinity
    17:39:46.077t1 === t2
    17:39:46.081true

  • Discourse touched me in a no-no place



  • @PJH said:

    http://ecma262-5.com/ELS5_HTML.htm#Section_11.9.6

    Ah, so that was by design. Two variables can be "exactly equal" but plugging them into a deterministic function yields two different results. I guess that's an anomaly to be filed next to "NaN !== NaN".


  • Discourse touched me in a no-no place

    @anotherusername said:

    I guess that's an anomaly to be filed next to "NaN !== NaN".
    That one's part of the fundamental definition of what NaN is.



  • @PJH said:

    That one's part of the fundamental definition of what NaN is.
    It's stupid. Just like NaN, Infinity, and -Infinity are all typeof "number" when they're actually not numbers. Also like people who write typeof x === "somesuch" instead of typeof x == "somesuch".


  • Discourse touched me in a no-no place

    @anotherusername said:

    @PJH said:
    That one's part of the fundamental definition of what NaN is.
    It's stupid. Just like NaN, Infinity, and -Infinity are all typeof "number" when they're actually not numbers. Also like people who write typeof x === "somesuch" instead of typeof x == "somesuch".
    You haven't been doing this programming lark for very long have you?



  • @PJH said:

    @anotherusername said:
    @PJH said:
    That one's part of the fundamental definition of what NaN is.
    It's stupid. Just like NaN, Infinity, and -Infinity are all typeof "number" when they're actually not numbers. Also like people who write typeof x === "somesuch" instead of typeof x == "somesuch".
    You haven't been doing this programming lark for very long have you?
    I'm curious how you can deduce that from what I've posted here.



  • @anotherusername said:

    T'm curious how you can deduce that from what I've posted here.
    I think it's because IEEE 754 and how it's implemented and thought about are generally well-known by most programmers who have had to do much arithmetic (read: every programmer) with using floating-point, since it tends to rear it's ugly head a lot. In addition, it's in every undergrad systems course, and has been more or less since its inception.

    @anotherusername said:

    @PJH said:
    That one's part of the fundamental definition of what NaN is.
    It's stupid. Just like NaN, Infinity, and -Infinity are all typeof "number" when they're actually not numbers. Also like people who write typeof x === "somesuch" instead of typeof x == "somesuch".


    To invoke some PL theory, you are given the task of writing the typing rule for the expression e1/e2, where against gamma e1 and e2 take on the type float. Without dependent types, or a dynamic type system, you can't choose a non-float type as a result iff the result of e1/e2 is NaN or Infinity. Also, infinity is a number.


  • Considered Harmful

    if( !isNaN( foo ) && isFinite( foo ) ) { // yawn



  • @lushr said:

    IEEE 754 and how it's implemented and thought about are generally well-known by most programmers who have had to do much arithmetic (read: every programmer) with using floating-point
    Knowing about it doesn't mean that I mustn't think it's completely stupid in some cases. And in general, I find it much easier to expect that it'll behave sanely without my delving in the details of the IEEE specs except in the bizarrely stupid cases where I've discovered it won't and I must.@lushr said:
    Without dependent types, or a dynamic type system
    So, in languages that aren't Javascript, which is dynamically typed.@lushr said:
    Also, infinity is a number.
    Not according to mathematics.



  • @Maciejasjmj said:

    They're just following the standards set by the Windows 8 BSOD:

     

    Wow, so that's what it looks like.

     



  • @anotherusername said:

    @lushr said:
    Also, infinity is a number.
    Not according to mathematics.
    What are these, then?



    Edit to add:

    Note that aleph 1 = |R| = |[-Infinity, Infinity]| = Infinity, so aleph 1 = infinity, at least to analysts.



  • @lushr said:

    @anotherusername said:
    @lushr said:
    Also, infinity is a number.
    Not according to mathematics.
    What are these, then?
    Something completely different:
    The aleph numbers differ from the infinity (∞) commonly found in algebra and calculus. Alephs measure the sizes of sets; infinity, on the other hand, is commonly defined as an extreme limit of the real number line (applied to a function or sequence that "diverges to infinity" or "increases without bound"), or an extreme point of the extended real number line.
    The infinity used in algebra and calculus is not a number. It is a limit.@lushr said:
    Note that aleph 1 = |R| = |[-Infinity, Infinity]| = Infinity, so aleph 1 = infinity, at least to analysts.

    You can't use infinity in the equation, because it isn't a number.

    You can use it as the endpoint for a set, but you should use open notation because infinity isn't a member of the set: (-Infinity, Infinity).

    [url]http://en.wikipedia.org/wiki/Interval_(mathematics)#Infinite_endpoints[/url]

    The notations [−∞, b] , [−∞, b) , [a, +∞] , and (a, +∞] are ambiguous. For authors who define intervals as subsets of the real numbers, those notations are either meaningless, or equivalent to the open variants.
    (I.e. (−∞, b] , (−∞, b) , [a, +∞) , and (a, +∞), respectively.)


  • @joe.edwards said:

    if( !isNaN( foo ) && isFinite( foo ) ) { // yawn

    Ask yourself: Is NaN a finite number? For that matter, is ""joe.edwards"" a finite number?



  • @Ben L. said:

    Ask yourself: Is NaN a finite number?

    NaN is Not A Number, but it is a Number primitive, but it's not a finite number. And if that wasn't confusing enough, just ask yourself, "Is 'Not A Number' a number, and if it was a number, would it be a finite number?"

    Personally, I think it'd be a purple number.


Log in to reply