Toshiba doesn't like my search terms



  • I was trying to find information about a NB500-100 netbook on toshiba-europe.com but it seems like the search engine doesn't want to search for that, it prefers to search for C660-1GD instead.

    If i remove the minus "-" from the search term it works as expected, so maybe the engine is smart and its subtraction 100 from NB500, and the result is obviously C660-1GD

    derp



  • Maybe this is an attempt to solve what it thinks is your algebra homework.

    As it's a "Search enhanced by Google" maybe it's part of a 20% time project.



  • Odd, this doesn't happen in a standard Google search here. I wonder where it's getting the rest from too. I have had a single instance in the past of it messing up; it evaluated some pare of the search query as Planck's constant which led to very strange results.



  •  "Search enhanced by Google" is becoming an oxymoron.  I had the throw-your-computer-out-the-window-rage-inducing moment of Googling "strrchr".  It "corrects" my search to strchr, then asks if I actually meant to search for what I typed in.  Yes, Google, thanks ever so much.  It condescends to search for my search term as I entered if and the first handful of results had no instances of "strrchr" on the entire page.  If Google are so keen on tracking every aspect of my behaviour online, why don't they twig that 80% of my searches are for programming language keywords and set some kind of "this user is not a drooling moron, search for the fucking search terms verbatim" flag?



  • @orange_robot said:

     80% of my searches are for programming language keywords

    @orange_robot said:

    this user is not a drooling moron

    I fail to see the connection


  • Discourse touched me in a no-no place

    @orange_robot said:

     "Search enhanced by Google" is becoming an oxymoron.  I had the throw-your-computer-out-the-window-rage-inducing moment of Googling "strrchr".  It "corrects" my search to strchr, then asks if I actually meant to search for what I typed in.
    WONTFIX: Cannot reproduce...



  • @serguey123 said:

    @orange_robot said:

     80% of my searches are for programming language keywords

    @orange_robot said:

    this user is not a drooling moron

    I fail to see the connection

    Considering in the last month, I've had to explain to posters on this very board what an API contract is, and also why a web browser should have a JavaScript debugger-- I'm starting to think "drooling morons" is generous for programmers.



  • @blakeyrat said:

    Considering in the last month, I've had to explain to posters on this very board what an API contract is, and also why a web browser should have a JavaScript debugger-- I'm starting to think "drooling morons" is generous for programmers.

    According to Google, there are all of maybe five relevant instances of the phrase "API contract" on the entire internet. So maybe it's you who is the drooling moron.

     

     



  • @Zylon said:

    @blakeyrat said:

    Considering in the last month, I've had to explain to posters on this very board what an API contract is, and also why a web browser should have a JavaScript debugger-- I'm starting to think "drooling morons" is generous for programmers.

    According to Google, there are all of maybe five relevant instances of the phrase "API contract" on the entire internet. So maybe it's you who is the drooling moron.

    Even though blackeyrat is very likely a high-functioning aspie and has the social skill of an angry four year old, he does seem to know his stuff, IT related anyway...



  • @C-Octothorpe said:

    he does seem to know his stuff, IT related anyway...
    Agreed - apart from the <sarcasm> tiny </sarcasm> percentage of IT "stuff" that involves Unix and/or anything open source - both of which blakerat makes it a solemn principle of his NOT to know anything about, for fear of getting cooties.



    (The sarcasm tags are not overkill considering the levels in this "community" of (a) autism and (b) preposterous underestimation of what that percentage actually is.)



  • My favourite is when you search for foo, and get an entire page of results that have absolutely nothing to do with foo. Then you notice a small link at the bottom of the page: "showing results for bar. Search instead for foo." No "did you mean" at the top - this one link at the bottom of the page is the only indication that Google has helpfully searched for something completely different. This doesn't happen every time, though.



  • Cool, I just googled "foo" and discovered there is an RFC for the etymology of "foo"!






    Here's the abstract:



    " Approximately 212 RFCs so far, starting with RFC 269, contain the

    terms foo', bar', or `foobar' as metasyntactic variables without

    any proper explanation or definition. This document rectifies that

    deficiency "



    Dirty job, but somebody's got to do it.



  • I'd say that RFC falls into the same category as this one: http://www.ietf.org/rfc/rfc1149.txt



  • If I had to guess, I'd say it came from FUBAR, which was used by British (or was it American) soldiers to mean "F***ed Up Beyond All Recognition"...



  • @token_woman said:

    @C-Octothorpe said:
    he does seem to know his stuff, IT related anyway...
    Agreed - apart from the <sarcasm> tiny </sarcasm> percentage of IT "stuff" that involves Unix and/or anything open source - both of which blakerat makes it a solemn principle of his NOT to know anything about, for fear of getting cooties.

    I know it, I just don't like it.

    @token_woman said:

    (The sarcasm tags are not overkill considering the levels in this "community" of (a) autism and (b) preposterous underestimation of what that percentage actually is.)

    Yes, a lot of IT stuff happens on *nix. Also, a lot of eating happens at McDonalds. That doesn't make *nix a good OS, and that doesn't make McDonalds a gourmet restaurant.

    If I'm going to spend my time writing a program, I'm going to write it for the OS that wasn't designed by clowns, that frequently and consistently has new features added, that has a stable installation system that works on every copy of the OS, that has a decent security system, that practically guarantees that a correctly-written program will keep running for the next 20 years without me having to give the source code away.



  • @blakeyrat said:

    I know it, I just don't like it.
    OK, I'm pretty sure I remember posts of yours claiming you don't know what "grep" means or whatever, but whatever.
    @blakeyrat said:
    If I'm going to spend my time writing a program, I'm going to write it for the OS that wasn't designed by clowns, that frequently and consistently has new features added, that has a stable installation system that works on every copy of the OS, that has a decent security system, that practically guarantees that a correctly-written program will keep running for the next 20 years without me having to give the source code away.
    Erm ... what?
    I'm genuinely confused here. Are you actually suggesting that Ubuntu (just to pick one) doesn't have new features added regularly? And that there is a currently-maintained OS on earth (with the exception of a few ploddy old extremely-reliable Unix beasts) which can run software from 1992? And that all the banks, military institutions and so on that use *nix are constantly having to give their source code away?

    As for the thing about installation systems working on every "copy of the OS", I don't understand what you mean - although it makes much more sense than the suggestion than Unix has no security (ah, those pesky banks and military institutions come to mind again).



  • @token_woman said:

    And that there is a currently-maintained OS on earth (with the exception of a few ploddy old extremely-reliable Unix beasts) which can run software from 1992?

    Yes. It's called "Microsoft Windows"



  • @token_woman said:

    OK, I'm pretty sure I remember posts of yours claiming you don't know what "grep" means or whatever, but whatever.

    Christ.

    Not everything I type is literally true. Let me append that: Nothing I type is literally true. Especially that last sentence.

    If you're going to be a pedantic dickweed and take everything I say as gospel, then you need to stop reading my posts, because you're just going to come away confused and I'm just going to get mad at you for being a pedantic dickweed.

    If my post would be funnier by my pretending I don't understand X (where X is "GREP" or whatever), then I pretend I don't understand X. If I can more easily make a point by pretending to not understand Y, then I pretend I don't understand Y. If I say statement Foo is true, even though I know that in borderline case Bar it might not be strictly true, I'm not going to spent 300 words discussing that pointless sidebar because it'll detract from my actual point, and Foo is generally accepted to be true anyway. If I use the word "I", I might not literally mean "myself".

    Take the following example:


    "Fixing video problems in Linux is easy, you just open the console and type: 'sudo dpkg-reconfigure xserver-xorg'."

    "Ok, you lost me at 'console'."

    See what I'm doing here is not literally saying that I don't understand what a console is. What I'm doing is putting myself in the shoes of a typical computer user (something most programmers are fucking godawful at, BTW, and Linux developers don't even fucking try) and pointing out that the instructions given, despite the claim, aren't "easy". Because they involve terms the average computer user does not, and should not, need to know.

    If I then say, "those instructions are so difficult, I'm going to adopt some puppies just so I can strangle them," I'm not literally going to the pound, going through the pointlessly bureaucratic adoption process to obtain puppies and then strangling them. What I actually do is break into a PetCo at 3:00 AM by throwing a brick through the window, steal the puppies, and drown them in a nearby lake. Got it? Good.

    Now with your newfound knowledge of these strange concepts like "literary devices", "humor", and "not being a gigantic douchenozzle who takes everything strictly literally, like Data from Star Trek but without the yellow skin", your homework assignment is to go back and read every post I've ever written and then come back here and apologize to me for your idiocy.



  • @blakeyrat said:

    If my post would be funnier by my pretending I don't understand X (where X is "GREP" or whatever), then I pretend I don't understand X.

    The other stuff you said, about security and what-not, showed far more ignorance than would not knowing about "grep". So to be charitable I'll assume you didn't mean any of it, and are just a silly old troll.



  • @Zylon said:

    @blakeyrat said:

    Considering in the last month, I've had to explain to posters on this very board what an API contract is, and also why a web browser should have a JavaScript debugger-- I'm starting to think "drooling morons" is generous for programmers.

    According to Google, there are all of maybe five relevant instances of the phrase "API contract" on the entire internet. So maybe it's you who is the drooling moron.

     

     

    You may find this search interesting: http://www.google.com/search?q=design+by+contract


  • @token_woman said:

    The other stuff you said, about security and what-not, showed far more ignorance than would not knowing about "grep". So to be charitable I'll assume you didn't mean any of it, and are just a silly old troll.

    I'm not knowledgeable in security or *nix, so I'm interested in your comments on this blog post, which details a hack to grant root access. Seems like this hack was found only because of *nix's open-source-ness. It seems to me that exposing source code is by nature less secure than not exposing it.

    There's a fix out already, but there are likely many, many systems that remain vulnerable to this because the companies that own them can't be bothered to recompile and redistribute the OS their applications run on.



  • @lettucemode said:

    Seems like this hack was found only because of *nix's open-source-ness. It seems to me that exposing source code is by nature less secure than not exposing it.

     

    Look, I'm hardly an Open Source zealot, but that's just idiotic. It's a paean to security through obscurity. You assume that 1) vulnerabilities won't be found in binary programs and 2) that finding and fixing vulnerabilities is a bad thing. It's completely out-of-touch with reality.



  • I've run Linux as my desktop OS for a decade now. I've only done serious development in the *nix world. I've been sysadmin for hundreds and hundreds of Linux servers, serving up enormous amounts of traffic.

     

    @token_woman said:

    Are you actually suggesting that Ubuntu (just to pick one) doesn't have new features added regularly?
     

    Blakeyrat was wrong here. Ubuntu adds new features all the time, it's just shit nobody wants. (Gnome3? Unity?)

     

    @token_woman said:

    And that there is a currently-maintained OS on earth (with the exception of a few ploddy old extremely-reliable Unix beasts) which can run software from 1992?

    Windows can do that, I believe.

     

    @token_woman said:

    And that all the banks, military institutions and so on that use *nix are constantly having to give their source code away?

    No, but if they wanted to allow someone to install their software in Linux, they're practically required to give away the source (or do custom binary builds for each target).

     

    @token_woman said:

    ..the suggestion than Unix has no security (ah, those pesky banks and military institutions come to mind again).

    I think he was being facetious here. Unix/Linux can be very secure, if you know what the fuck you are doing. The thing is, most people (even sysadmins) don't know what the fuck they are doing. Windows has better security out-of-the-box.

     



  • @morbiuswilters said:

    Look, I'm hardly an Open Source zealot, but that's just idiotic. It's a paean to security through obscurity. You assume that 1) vulnerabilities won't be found in binary programs and 2) that finding and fixing vulnerabilities is a bad thing. It's completely out-of-touch with reality.

    I'll clarify. I assume that finding vulnerabilities in binary programs is much more difficult than finding vulnerabilities in C programs, and anything you can do to raise the bar for entry into your system is a good thing. Though I recognize that Linux prefers its open-source-ness to obfuscating its source code, such an action lowers the bar for malicious entry.


  • ♿ (Parody)

    @lettucemode said:

    I'll clarify. I assume that finding vulnerabilities in binary programs is much more difficult than finding vulnerabilities in C programs, and anything you can do to raise the bar for entry into your system by the malicious types is a good thing. Then again, maybe Linux doesn't really care about the malicious types since they don't bother with Linux very much.

    There are automated tools for finding vulnerabilities with binary programs. You really don't need to know much of anything about how the target program was written. It's hard enough to read my own source code after a few months, let alone get inside the head of someone else. Granted, some things can be found with static analysis tools, but that doesn't seem to be the case with that particular vulnerability.



  • @lettucemode said:

    I'll clarify. I assume that finding vulnerabilities in binary programs is much more difficult than finding vulnerabilities in C programs, and anything you can do to raise the bar for entry into your system is a good thing. Though I recognize that Linux prefers its open-source-ness to obfuscating its source code, such an action lowers the bar for malicious entry.

    I know what you are saying, and you are wrong. Binary software is not terribly difficult to analyze. You'd think the sheer quantity of Windows and OSX 0 days would attest to that.

     

    What you are proposing is security through obscurity. In crypto, it's the idea that "keeping our cipher algorithm secret makes it more secure" which is bullshit. For one thing, discovering the algorithm is not difficult. Primarily, though, a well-written algorithm needs to be extremely resistant to cryptanalysis. If it does that, keeping the algorithm secret raises the bar for malicious entry by such a tiny amount it is meaningless. If you algorithm is vulnerable, then keeping it secret won't do jack shit.

     

    I'm not suggesting the opposite, either, that by making something open source it will suddenly be more secure. What I am saying is that if keeping your source code secret forms an integral part of your security posture, then you are doing it fucking wrong.



  • @morbiuswilters said:

    @lettucemode said:

    I'll clarify. I assume that finding vulnerabilities in binary programs is much more difficult than finding vulnerabilities in C programs, and anything you can do to raise the bar for entry into your system is a good thing. Though I recognize that Linux prefers its open-source-ness to obfuscating its source code, such an action lowers the bar for malicious entry.

    I know what you are saying, and you are wrong. Binary software is not terribly difficult to analyze. You'd think the sheer quantity of Windows and OSX 0 days would attest to that.

     

    What you are proposing is security through obscurity. In crypto, it's the idea that "keeping our cipher algorithm secret makes it more secure" which is bullshit. For one thing, discovering the algorithm is not difficult. Primarily, though, a well-written algorithm needs to be extremely resistant to cryptanalysis. If it does that, keeping the algorithm secret raises the bar for malicious entry by such a tiny amount it is meaningless. If you algorithm is vulnerable, then keeping it secret won't do jack shit.

     

    I'm not suggesting the opposite, either, that by making something open source it will suddenly be more secure. What I am saying is that if keeping your source code secret forms an integral part of your security posture, then you are doing it fucking wrong.

    OK, I defer to your experience. Thanks for the write-up.



  • @lettucemode said:

    @morbiuswilters said:

    @lettucemode said:

    I'll clarify. I assume that finding vulnerabilities in binary programs is much more difficult than finding vulnerabilities in C programs, and anything you can do to raise the bar for entry into your system is a good thing. Though I recognize that Linux prefers its open-source-ness to obfuscating its source code, such an action lowers the bar for malicious entry.

    I know what you are saying, and you are wrong. Binary software is not terribly difficult to analyze. You'd think the sheer quantity of Windows and OSX 0 days would attest to that.

     

    What you are proposing is security through obscurity. In crypto, it's the idea that "keeping our cipher algorithm secret makes it more secure" which is bullshit. For one thing, discovering the algorithm is not difficult. Primarily, though, a well-written algorithm needs to be extremely resistant to cryptanalysis. If it does that, keeping the algorithm secret raises the bar for malicious entry by such a tiny amount it is meaningless. If you algorithm is vulnerable, then keeping it secret won't do jack shit.

     

    I'm not suggesting the opposite, either, that by making something open source it will suddenly be more secure. What I am saying is that if keeping your source code secret forms an integral part of your security posture, then you are doing it fucking wrong.

    OK, I defer to your experience. Thanks for the write-up.

    No problem, dude. Sorry if I came off a bit rude, but I'm just so tired of the open source vs. closed source debates. Both sides produce good software and both sides produce a shit-ton of bad software.



  • @morbiuswilters said:

    I've run Linux as my desktop OS for a decade now. I've only done serious development in the *nix world. I've been sysadmin for hundreds and hundreds of Linux servers, serving up enormous amounts of traffic.
     

    Your cock grew three sizes typing that, did't it.



  • @blakeyrat said:

    If my post would be funnier by my pretending I don't understand X (where X is "GREP" or whatever), then I pretend I don't understand X.
     

    What a delightfully convenient metaphorical get-out-of-jail-free card. Genius!

     

     



  • @dhromed said:

    @morbiuswilters said:

    I've run Linux as my desktop OS for a decade now. I've only done serious development in the *nix world. I've been sysadmin for hundreds and hundreds of Linux servers, serving up enormous amounts of traffic.
     

    Your cock grew three sizes typing that, did't it.

    Yes, now I can almost touch both sides of the soda bottle neck.



  • @morbiuswilters said:

    No problem, dude. Sorry if I came off a bit rude, but I'm just so tired of the open source vs. closed source debates. Both sides produce good software and both sides produce a shit-ton of bad software.

    I don't have a problem with rudeness on a forum, especially when trying to convince someone of something (obligatory XKCD link), so you're fine. Sorry if I came off as dumb and uninformed, because I was :D



  • @lettucemode said:

    @morbiuswilters said:
    No problem, dude. Sorry if I came off a bit rude, but I'm just so tired of the open source vs. closed source debates. Both sides produce good software and both sides produce a shit-ton of bad software.

    I don't have a problem with rudeness on a forum, especially when trying to convince someone of something (obligatory XKCD link), so you're fine. Sorry if I came off as dumb and uninformed, because I was :D

    I don't think you came off as dumb or even all that uninformed. You had a position that many people take (that open sourcing software makes it less secure or more secure) but one I strongly feel is incorrect.

    You're pretty new here, long-time lurker or did you just discover the site or forums?



  • I found the website about a year ago, discovered the forums a few months ago, made an account this past December. One of my favorites sites for sure.


  • Trolleybus Mechanic

    @blakeyrat said:

    If I then say, "those instructions are so difficult, I'm going to adopt some puppies just so I can strangle them," I'm not literally going to the pound, going through the pointlessly bureaucratic adoption process to obtain puppies and then strangling them. What I actually do is break into a PetCo at 3:00 AM by throwing a brick through the window, steal the puppies, and drown them in a nearby lake. Got it? Good.
     

    Fucking figures. Of course you would turn to a commercial provider of closed source puppies. If you got a *REAL* open source puppy, you could just breed as many other puppies as you wanted, and strangle to suite your personal needs.




  • @Lorne Kates said:

    @blakeyrat said:

    If I then say, "those instructions are so difficult, I'm going to adopt some puppies just so I can strangle them," I'm not literally going to the pound, going through the pointlessly bureaucratic adoption process to obtain puppies and then strangling them. What I actually do is break into a PetCo at 3:00 AM by throwing a brick through the window, steal the puppies, and drown them in a nearby lake. Got it? Good.
     

    Fucking figures. Of course you would turn to a commercial provider of closed source puppies. If you got a REAL open source puppy, you could just breed as many other puppies as you wanted, and strangle to suite your personal needs.


    And he avoids vendor lock-in: he's got access to the puppy DNA, all he has to do is fork it and he can modify it to his heart's content.

    Of course, he's got to be careful with his licenses. Puppies are licensed under the GPL but rope is dual-licensed under a proprietary commercial license and the MIT license for non-commercial use. He's likely not going to make any money off of this, so he should be okay with the MIT license. However, the MIT license isn't GPL-compatible, so rope and puppy can't be linked.

    However, there is a Puppy Toolkit (PTK) that's released under the less-restrictive LGPL. It gives him 90% of the pieces he needs to build his own puppy; all that's missing is the head (although head support should be added in 0.0.99-BETA3).

    Strangling a puppy without a head isn't going to be very satisfying, though, so I recommend using BPHALXWM (Better Puppy Head And Lightweight X-11 Window Manager); just clone their git repo. There's no API, but if you know C you can easily graft BPHALXWM's puppy head onto your PTK puppy body. He doesn't need the X-11 Window Manager part of BPHALXWM (although it is a really great, lightweight window manager--there's absolutely no documentation to bloat the file size and you can remap keys if you know x64 assembly). The X-11 Window Manager itself is small but it links some dynamic libraries that aren't going to be available on his system, so I recommend editing the Makefile to skip building that part.



  • @morbiuswilters said:

    @Lorne Kates said:

    @blakeyrat said:

    If I then say, "those instructions are so difficult, I'm going to adopt some puppies just so I can strangle them," I'm not literally going to the pound, going through the pointlessly bureaucratic adoption process to obtain puppies and then strangling them. What I actually do is break into a PetCo at 3:00 AM by throwing a brick through the window, steal the puppies, and drown them in a nearby lake. Got it? Good.
     

    Fucking figures. Of course you would turn to a commercial provider of closed source puppies. If you got a REAL open source puppy, you could just breed as many other puppies as you wanted, and strangle to suite your personal needs.


    And he avoids vendor lock-in: he's got access to the puppy DNA, all he has to do is fork it and he can modify it to his heart's content.

    Of course, he's got to be careful with his licenses. Puppies are licensed under the GPL but rope is dual-licensed under a proprietary commercial license and the MIT license for non-commercial use. He's likely not going to make any money off of this, so he should be okay with the MIT license. However, the MIT license isn't GPL-compatible, so rope and puppy can't be linked.

    However, there is a Puppy Toolkit (PTK) that's released under the less-restrictive LGPL. It gives him 90% of the pieces he needs to build his own puppy; all that's missing is the head (although head support should be added in 0.0.99-BETA3).

    Strangling a puppy without a head isn't going to be very satisfying, though, so I recommend using BPHALXWM (Better Puppy Head And Lightweight X-11 Window Manager). There's no API, but if you know C you can easily graft BPHALXWM's puppy head onto your PTK puppy body. He doesn't need the X-11 Window Manager part of BPHALXWM (although it is a really great, lightweight window manager--there's absolutely no documentation to bloat the file size and you can remap keys if you know X64 assembly). The X-11 Window Manager itself is small but it links some dynamic libraries that aren't going to be available on his system, so I recommend editing the Makefile to skip building that part.

    dunno man, seems complicated



  • @dhromed said:

    @morbiuswilters said:

    I've run Linux as my desktop OS for a decade now. I've only done serious development in the *nix world. I've been sysadmin for hundreds and hundreds of Linux servers, serving up enormous amounts of traffic.
     

    Your cock grew three sizes typing that, did't it.

    You mean that's not morbiuswilters in the picture?



  • @token_woman said:

    @dhromed said:

    @morbiuswilters said:

    I've run Linux as my desktop OS for a decade now. I've only done serious development in the *nix world. I've been sysadmin for hundreds and hundreds of Linux servers, serving up enormous amounts of traffic.
     

    Your cock grew three sizes typing that, did't it.

    You mean that's not morbiuswilters in the picture?
    There are plenty of Asian "women" working in massage parlors that look at least that feminine and also have cocks.


  • @token woman said:

    You mean that's not morbiuswilters in the picture?
     

    Yes it is. Morbs is an obscure internet-memetic asian chick.



  • @Salamander said:

    @token_woman said:
    And that there is a currently-maintained OS on earth (with the exception of a few ploddy old extremely-reliable Unix beasts) which can run software from 1992?

    Yes. It's called "Microsoft Windows"




    z/OS wins, it is currently-maintained system that can run binary software that was compiled in the 1960s...



    ... and that fact is the reason Fortune 500 companies still pay millions for IBM mainframes, instead of rewriting.



  • @frits said:

    @token_woman said:
    You mean that's not morbiuswilters in the picture?
    There are plenty of Asian "women" working in massage parlors that look at least that feminine and also have cocks.

    You seem to imply that somewhere there is a kind of people that look a lot like a woman but also run Linux and are not a freetard / zealot. Where do I sign?

    Oh drat, the ink evaporated.



  • To return for a moment to the original WTF:

    @BaRRaKID said:

    I was trying to find information about a NB500-100 netbook on toshiba-europe.com but it seems like the search engine doesn't want to search for that, it prefers to search for C660-1GD instead.

    If i remove the minus "-" from the search term it works as expected, so maybe the engine is smart and its subtraction 100 from NB500, and the result is obviously C660-1GD

     

    This would not be an issue if not for the fact that people can't tell the difference between a hyphen and a minus sign.

     



  • @da Doctah said:

    This would not be an issue if not for the fact that people can't tell the difference between a hyphen and a minus sign.

    One says you lost your virginity and the other says you lost at poker.. right?

    Disclaimer: I am not a Doctah.



  • @orange_robot said:

    I had the throw-your-computer-out-the-window-rage-inducing moment of Googling "strrchr".  It "corrects" my search to strchr, then asks if I actually meant to search for what I typed in.  Yes, Google, thanks ever so much.  It condescends to search for my search term as I entered if and the first handful of results had no instances of "strrchr" on the entire page.

    Are you absolutely, positively certain that you didn't put in an extra r and accidentally search for strrrchr instead? Because "strrchr" works for me, but "strrrchr" auto-corrects to strchr. So does "strrchrr". Though with any variant I've tried, if you click the "yes, search for what I typed link" the top results all demonstrate strrchr.

    And I'm not a drooling moron either, and generally know what I'm doing, yet I still make the occasional typo which I don't always catch immediately. If Google corrects it for me, that's a good thing. And that happens much more frequently than the "no, I really did mean what I typed and not some other much more popular term" issue.

    For the OP: @BaRRaKID said:

    If i remove the minus "-" from the search term it works as expected, so maybe the engine is smart and its subtraction 100 from NB500, and the result is obviously C660-1GD
    Much more likely it's interpreting the request as "find pages with 'NB500', excluding pages with '100'". Then it can't find any because all the NB500 pages have NB500-100 in them, so it returns the closest thing it can find.

Log in to reply