Ridiculous input validation gone bad (for the some of the world)



  • tried to opt-out of their newsletter.

    input validation to ensure you don't copy-paste your email (why not actually, to begin with???) is catch the use of ctrl and alt keys.

    wtf #1: to type the character '@' i have to use the 'alt gr' key on my keyboard

    wtf #2: i can paste easily using shift-insert

     

    hence: not only annoying, but also doesn't achieve the goal...!

     

    yeah, yeah, TRWTF is you need the 'alt gr' key on sg keyboard layout to get to the vital '@' char :-((

     screenshot

     



  • @IronFist said:

    yeah, yeah, TRWTF is you need the 'alt gr' key on sg keyboard layout to get to the vital '@' char :-((

    On all keyboards layouts I have seen (French PC, Swiss PC and Mac), you need the 'alt', 'ctrl' or 'alt gr' key for "special" characters (like @ or |[]{}# on mac)

    But seems on US keyboards you need 'Shift+2'

    Restricting keyboard usage is a WTF.

    Solution: Disable javascript ... and wait ... no form submit also in javascript



  • Well, it's okay. After all, only Americans use British Telecom, right?



  • @ltouroumov said:

    On all keyboards layouts I have seen (French PC, Swiss PC and Mac), you need the 'alt', 'ctrl' or 'alt gr' key for "special" characters (like @ or |[{}# on mac)

    But seems on US keyboards you need 'Shift+2'

     

    The standard US keyboard layout does not use Ctrl, Alt, or Alt Gr to enter any symbols. There isn't actually an Alt Gr key on the standard layout, just a second Alt key. The US International layout has an Alt Gr key that you use to enter foreign currency symbols and some common accented letters.



  • @MiffTheFox said:

    Well, it's okay. After all, only Americans use British Telecom, right?

    The network administration of a major US manufacturer I have done work with in South Carolina, is outsourced to BT. Not only that, but to the Caribbean office of BT!



  • Can't believe nobody's said it yet...

    TRWTF is BT.

    (Example: a couple of weeks ago, BT came to do a site survey - checking that the fibre they needed to use for a new circuit was in order, basically. Today, they came to do some fibre splicing. And decided that the fibre they needed wasn't there, so they'd have to send someone to do a site survey. I expect this to go in circles for months....)

    Oh, and on UK Windows keyboards, you just need a shift key to get an @. Which is probably why they didn't notice they were creating a problem

    On the other hand, it might be a deliberate act, and part of their overall Evil Plan.



  • @Someone You Know said:

    The standard US keyboard layout does not use Ctrl, Alt, or Alt Gr to enter any symbols. There isn't actually an Alt Gr key on the standard layout, just a second Alt key. The US International layout has an Alt Gr key that you use to enter foreign currency symbols and some common accented letters.

    My Acer netbook (purchased in the US from Newegg) has the second alt key marked as “Alt Gr” but it doesn't actually function as Alt Gr unless I load my custom Window-1252 layout. It’s the only keyboard I’ve actually seen with an Alt Gr, except for the Danish keyboard we have from when we had a Danish programmer visiting.



  • All UK keyboards have had an AltGr key for at least ten years; basically ever since the Euro came into being. On a UK keyboard, you press AltGr+4 for a € symbol. And you press AltGr+` to get a ¦ symbol. Those symbols are both usually marked on those keys, typically causing much head-scratching from lusers who can SEE the € symbol there, but have no clue which key(s) to press to make one appear on the screen.



  • Is THAT how you get to it? I've seriously been wondering all this time because it looks like it should be a shift-4 but that of course spawns the $ symbol.

    Interestingly, this two month old UK HP G56 doesn't have a Euro symbol. Though the entire top row of keys is shunted one to the left by a dedicated lock button and the resulting gap down the left hand side populated by fixed-function macro-based shortcut keys so I doubt the designers had a whole lot of common sense. Putting a dedicated Outlook button below ESC, for instance. Or a dedicated print key left of shift. At least the calculator button, if in a silly place left of CTRL, gets a bit of use.



  • Heh, when I had to restrict copy/paste in the internal app I made, I just counted keypresses. If the number of keypresses (backspace and delete decrement the counter ofc) is different from the length of the string, you get error.

    As for AltGr I use it all the time as well, all kinds of special characters I need are accessible for me only with AltGr: @{}<>[]


  • Garbage Person

     About half my stash of Model M's (US 101key) have Alt Gr marked. IBM had the brilliant idea of making the text green (and on the rare keyboard that actually has the AltGr characters shown - I have a total of one Model M with such), those characters are also printed in green. Of course, I use a Mk2 Das Keyboard on my primary machine - I don't need no stinking printed keycaps.



  • Heh this reminds me of these ridiculous right-click intercepting scripts that try to prevent you from downloading images by fucking with the context menu. First time I saw them was in 2000, and there were not only Javascript but also ActiveX flavors that only worked in IE (back when executing random binary content was considered typical internet usage).

    Needless to say, they were utterly useless and annoying. Even so, you can still see them used here and there. "Our images are copyright protected" - wow, really? Even assuming that the images are indeed marketed as such and ARE commodities under the Law, why would you restrict downloading the web-resolution preview, when you have a full-resolution marketable to back it up? And you [i]do[/i], right? Trying to protect any [i]full[/i] content you place for public download [i]a posteriori[/i] is TRWTF.



  • @Weng said:

     About half my stash of Model M's (US 101key) have Alt Gr marked. IBM had the brilliant idea of making the text green (and on the rare keyboard that actually has the AltGr characters shown - I have a total of one Model M with such), those characters are also printed in green. Of course, I use a Mk2 Das Keyboard on my primary machine - I don't need no stinking printed keycaps.

     

    So it's short for Alt Green, then.



  • @Shinhan said:

    Heh, when I had to restrict copy/paste in the internal app I made, I just counted keypresses. If the number of keypresses (backspace and delete decrement the counter ofc) is different from the length of the string, you get error.

    As for AltGr I use it all the time as well, all kinds of special characters I need are accessible for me only with AltGr: @{}<>[

    I always find it compelling when people tell these stories, with no apparent shame, of how they decreased the usability of their app.



  • @ltouroumov said:

    On all keyboards layouts I have seen (French PC, Swiss PC and Mac), you need the 'alt', 'ctrl' or 'alt gr' key for "special" characters (like @ or |[{}# on mac)

    But seems on US keyboards you need 'Shift+2'

    On the keyboards I use I also need Alt+Gr for those essential characters (except | which is Shift+\ just below Esc; and # which is Shift+3).

    But the real reason I'm quoting you is because I'm reminded of how, back in college, a fellow student / friend of mine helped another CompSci student get to the @ character. It seems on the old US keyboards of the monochrome Unix terminals we had, you had to simultaneously press Shift and 2 to get to the character engraved above 2.

    Yes, she was blonde.*

    @C4I_Officer said:

    Heh this reminds me of these ridiculous right-click intercepting scripts that try to prevent you from downloading images by fucking with the context menu.
    Again back in college, there was this page which prevented you from using the context menu (I can't remember any more details other than this).
    First time you tried, it warned you with a friendly alert saying it would crash your browser if you insisted.
    The second time, it would run function overload() { overload; } - which succeed to crash IE4.

     

    * TRWTF is we actually had a real life blonde hot girl taking CS.



  • @Weng said:

    About half my stash of Model M's (US 101key) have Alt Gr marked … Of course, I use a Mk2 Das Keyboard on my primary machine - I don't need no stinking printed keycaps.

    It's great to see that we have at least one quality keyboard enthusiast here! I use FILCOs at the moment, one being Cherry MX blue the same as your Das. Personally I prefer lettered keyboards for aesthetic reasons; even Star Trek looked daft with unlabelled controls.

    @Zecc said:

    First time you tried, it warned you with a friendly alert saying it would crash your browser if you insisted.

    The second time, it would run function overload() { overload; } - which succeed to crash IE4.

    That's only going to return a function pointer! You want: function overload() { overload(); }

    It doesn't work in IE 7 though, it triggers "Windows Internet Explorer: Stack overflow at line: 13" instantly. However, it's not too hard to crash Internet Explorer 7 and earlier: just go to crashie.com (oddly I can't get what appears to be the fatal code to work on a local page). There was another site years ago called something like crashmypc.com that opened lots of Flash-based windows until the system fell over; the only system I couldn't get it to crash was Mac OS 8.5 – it took out Windows NT 4 and Linux Mandrake easily. The Mac's memory partition system led to the browser's termination long before its memory usage overran the OS. Sadly that site no longer exists.

    @Zecc said:

    * TRWTF is we actually had a real life blonde hot girl taking CS.

    Did she graduate though?



  • Oh dear. crashie.com still works in Internet Explorer 8 – instant hang (the site only claims that it works on 7 and earlier, which I foolishly took to mean that Microsoft had actually fixed the bug in 8). Very reassuring.

    Opera blocks it as a malware site, which is funny as it has no effect in Opera at all.



  • @Daniel Beardsmore said:

    Oh dear. crashie.com still works in Internet Explorer 8 – instant hang

    No it doesn't. The error isn't particularly useful, but it's no crash:

    @IE8 said:

    Webpage error details

    User Agent: Mozilla/4.0 (compatible; MSIE 8.0; Windows NT 6.1; WOW64; Trident/4.0; SLCC2; .NET CLR 2.0.50727; .NET CLR 3.5.30729; .NET CLR 3.0.30729; Media Center PC 6.0; .NET4.0C; .NET4.0E)
    Timestamp: Sun, 20 Feb 2011 16:18:27 UTC

    Message: Unspecified error.
    Line: 1
    Char: 52
    Code: 0
    URI: http://www.crashie.com/

    Message: Unspecified error.
    Line: 1
    Char: 52
    Code: 0
    URI: http://www.crashie.com/

    Anyway, it kind of falls flat when IE was the first browser to introduce security features like address space randomization, sandboxing, malware site database, anti-phishing filter, IDN highlighting, ActiveX running in a separate process, etc. Gripe about IE being slow to introduce new features, or having poor performance, but security-wise it's pretty damned solid since IE7. They also invented the little yellow bar block-by-default security mechanism which every other browser now has copied.

    @Daniel Beardsmore said:

    (the site only claims that it works on 7 and earlier, which I foolishly took to mean that Microsoft had actually fixed the bug in 8).

    The code they're using to "crash" it is: <script>for(x in document.write){document.write(x);}</script>

    "Fix" it to do what? It's complete nonsense... what does "for( x in ANativeCodeFunction) even supposed to do? I suppose it could run one loop and return the function's toString()... which would just be "[native code]". (I mean, not that failing on nonsense code is a defense against crashing, but it's not like this code is supposed to "do" anything at all.)

    @Daniel Beardsmore said:

    Opera blocks it as a malware site, which is funny as it has no effect in Opera at all.

    Well, it is a malware site. So there's that. Secondly, Opera's malware list is probably coming from a browser-agnostic third-party, so they won't necessarily have a "site is malware but doesn't affect Opera" boolean in it.



  • @Daniel Beardsmore said:

    That's only going to return a function pointer! You want: function overload() { overload(); }
    Yeah, I'm blonde too and momentarily forgot to press Shift+8,9.@Daniel Beardsmore said:
    Did she graduate though?
    I'm not sure, but I don't remember seeing her in the latter years, no. I prefer brunettes anyway.



  • @blakeyrat said:

    Long, worthless rant indicating total absence of sense of humour

    Yay, my personal humourless naysayer is back.

    I definitely get a crash in IE 8 identical to 7; maybe this is like how XP got a superior IE 6 to Windows 2000's IE 6 – I'm still in XP. Part of me would love to go 7, the rest of me is content to wait until I have no choice, because the only real improvement seems to be that Explorer errors no longer terminate a copy/move operation, but it's staggering that Microsoft took until Windows Vista to put a Retry button into the shell.

    I don't gripe about IE 8's performance, though opening a block of new tabs is tedious, but I hear a lot of complaints about IE 8's performance. Doesn't surprise me how many people are moving to Firefox and Chrome to get something that performs better. Personally, my gripe with IE remains that it's clunky, uninspired, and ill-designed. I am not that much of masochist.



  • @Daniel Beardsmore said:

    Yay, my personal humourless naysayer is back.

    OH NOES I R HUMORLESS NAYSAYER!!!

    I just don't like to see people piling-on IE for no good reason. Partially because it's not that bad a product by any stretch of the imagination, but mostly because it's such a fucking annoying cliche... it's worse than quoting Monty Python. How about we gripe about products that are *actually* shit?

    @Daniel Beardsmore said:

    Part of me would love to go 7, the rest of me is content to wait until I have no choice, because the only real improvement seems to be that Explorer errors no longer terminate a copy/move operation, but it's staggering that Microsoft took until Windows Vista to put a Retry button into the shell.

    That's seriously the only improvement you can think of between XP and Windows 7? ... seriously? Are you... trolling in some obscure way?

    Ok, you're officially unqualified to comment on anything Vista or Windows 7 related, since you're obviously supremely ignorant of those products.

    @Daniel Beardsmore said:

    I don't gripe about IE 8's performance, though opening a block of new tabs is tedious, but I hear a lot of complaints about IE 8's performance.

    IE does't look as bad now that Firefox has bloated-up. And before Chrome had feature-equivalence, at least Microsoft could argue that IE did more. Now... now it's just kind of sluggish. Oh well.

    @Daniel Beardsmore said:

    Personally, my gripe with IE remains that it's clunky, uninspired, and ill-designed.

    Well, it was designed well enough to destroy Netscape. (Of course, Netscape worked pretty hard to destroy themselves, so.) It's inspired all other browsers to rip-off the IE7 UI-- not just the yellow bar for security issues, but the general arrangement of tabs and menus.

    I won't argue with clunky, but, again: IE is not nearly, not even in the same order of magnitude, as bad a product as people like you think it is.

    But since you honestly believe Windows 7 has no improvements over XP (except the file copy dialog) you've actually just placed yourself in the category "Slashdot-esque tech Luddite!" Congratulations! This signifies that you have lost all ability to comparatively judge products, have absolutely no sense of curiosity, and moreover, have no interest in advancing the industry! Why, next thing you know you'll be bitching and whining that you can't get a cellphone without a camera in it, or spending hours reminiscing about how great WordPrefect 5.1 was.



  • IE has come on a lot over years, granted; it's nice to feel relatively assured that CSS and JavaScript developed in Firefox will work in IE 8. However, until such time that it ever becomes pleasant to use, its praises won't get sung. And if IE remains perpetually behind the curve in both standards support and usability, it's going to have a hard job improving its reputation. What you must remember is that anything good in the UI, Firefox got there a long time before, and iCab or Opera probably did before Firefox. So even when Microsoft do get something halfway right, anyone with a clue was using something better long before that. IE doesn't just have to catch up, it has to overtake.

    Am I trolling? If I am, then you took the bait very predictably, and again, that says more about you than it does about me. The real deal for me is whether I return from using Vista/7/2008/2008 R2 to XP and think to myself, "self, XP really sucks at <something> where 7 succeeds". 7 lets you drag-rearrange the taskbar and tray icons, but I was doing this in XP long before 7 was released. A program called Everything lets me find any file by name instantly.

    The problem with any OS overhaul is that you're always back where you started. 7's Explorer has no commands toolbar and no tasks panel (that I'm aware of) so I can't imagine how anyone non-technical (parents for example) could seriously use it like that; don't ask me why context menus are so hard to understand, but for some reason they are. 7 has no "elevate" command at the command line, and a lot of the UI is reduced to random invocations involving keyboard modifiers, e.g. command-shift-click on taskbar button to open a new instance with elevation.

    The new taskbar is a confusing mess because it relies too much on an impossible state: every program being designed properly. You've got programs that can't be pinned because the taskbar can't decide what to invoke to open them, MMC snap-ins that all show up as the same icon so you can't tell one from another, programs that won't elevate with ctrl-shift-click, windows that are still omitted from the taskbar for no reason …

    You need a third-party hack to force shares to get their share badge shown on the icon again, and pretty much, 7 is back to collecting a fresh new set of toys to get it to behave sensibly, where with XP I've already gone that and got settled into something that already works pretty much as well as 7 does.

    How do limited users install fonts in 7? In XP, you can relax the security on C:\WINDOWS\Fonts so that limited users can install fonts, but in 7, C:\WINDOWS\Fonts is hacked up so badly that it doesn't even have a Security tab in Properties and cacls is also banned from altering permissions. Why can't Windows, by 7, have per-user fonts like the Macintosh?

    I don't desire to hate Windows 7, or anything else for that matter; I'm just aware that it has years to go before it's a fully consistent environment, and I'm not swayed by shadows, translucency, pointless animations and other fluff. I was using a Sony VAIO 1920×1080 15.5″ recently, set to 125% DPI, and Microsoft still don't have a clue how to scale graphics. They still don't understand that to scale an 8-bit image, it must be raised to 32-bit first, and there are a lot of problems in their scaling in general that leaves the system looking very rough-edged. Again, in time, high-DPI will be the dream come true that it should be, but right now, it looks like a hack. Another thing is that Windows 7 still retains the shoddy ClearType from XP, and only WPF gets the more advanced version with sub-pixel positioning. This means that in 125% high-DPI mode, fonts that were OK with 1px strokes *still* have 1px strokes, so text looks quite spindly!

    I was sure that I'd been told, by people here, that Windows 7 introduced revised ClearType, and wondered why ClearType still sucked on every Windows 7 machine I used. I've since learnt that only WPF gets that, so who knows how many years it will be before all Windows programs move to WPF and finally get smooth text. I guess never, because we'll have 1200 DPI displays long before then and the argument will become moot. Hell, at high DPI even regular antialiasing will be perfectly adequate, no need for ClearType in the first place, which is a win because ClearType is an awful hack really. (The fun phase will be screenshots taken from 300 DPI machines being shown on 100 DPI screens …)



  • @Daniel Beardsmore said:

    (Pointless Ranting)

    Yah. I didn't read that. Just so you know.



  • @Daniel Beardsmore said:

    @blakeyrat said:

    Long, worthless rant indicating total absence of sense of humour

    @Daniel Beardsmore said:

    IE has come on a lot over years, granted; it's nice to feel relatively assured that CSS and JavaScript developed in Firefox will work in IE 8. However, until such time that it ever becomes pleasant to use, its praises won't get sung. And if IE remains perpetually behind the curve in both standards support and usability, it's going to have a hard job improving its reputation. What you must remember is that anything good in the UI, Firefox got there a long time before, and iCab or Opera probably did before Firefox. So even when Microsoft do get something halfway right, anyone with a clue was using something better long before that. IE doesn't just have to catch up, it has to overtake.

    Am I trolling? If I am, then you took the bait very predictably, and again, that says more about you than it does about me. The real deal for me is whether I return from using Vista/7/2008/2008 R2 to XP and think to myself, "self, XP really sucks at <something> where 7 succeeds". 7 lets you drag-rearrange the taskbar and tray icons, but I was doing this in XP long before 7 was released. A program called Everything lets me find any file by name instantly.

    The problem with any OS overhaul is that you're always back where you started. 7's Explorer has no commands toolbar and no tasks panel (that I'm aware of) so I can't imagine how anyone non-technical (parents for example) could seriously use it like that; don't ask me why context menus are so hard to understand, but for some reason they are. 7 has no "elevate" command at the command line, and a lot of the UI is reduced to random invocations involving keyboard modifiers, e.g. command-shift-click on taskbar button to open a new instance with elevation.

    The new taskbar is a confusing mess because it relies too much on an impossible state: every program being designed properly. You've got programs that can't be pinned because the taskbar can't decide what to invoke to open them, MMC snap-ins that all show up as the same icon so you can't tell one from another, programs that won't elevate with ctrl-shift-click, windows that are still omitted from the taskbar for no reason …

    You need a third-party hack to force shares to get their share badge shown on the icon again, and pretty much, 7 is back to collecting a fresh new set of toys to get it to behave sensibly, where with XP I've already gone that and got settled into something that already works pretty much as well as 7 does.

    How do limited users install fonts in 7? In XP, you can relax the security on C:\WINDOWS\Fonts so that limited users can install fonts, but in 7, C:\WINDOWS\Fonts is hacked up so badly that it doesn't even have a Security tab in Properties and cacls is also banned from altering permissions. Why can't Windows, by 7, have per-user fonts like the Macintosh?

    I don't desire to hate Windows 7, or anything else for that matter; I'm just aware that it has years to go before it's a fully consistent environment, and I'm not swayed by shadows, translucency, pointless animations and other fluff. I was using a Sony VAIO 1920×1080 15.5″ recently, set to 125% DPI, and Microsoft still don't have a clue how to scale graphics. They still don't understand that to scale an 8-bit image, it must be raised to 32-bit first, and there are a lot of problems in their scaling in general that leaves the system looking very rough-edged. Again, in time, high-DPI will be the dream come true that it should be, but right now, it looks like a hack. Another thing is that Windows 7 still retains the shoddy ClearType from XP, and only WPF gets the more advanced version with sub-pixel positioning. This means that in 125% high-DPI mode, fonts that were OK with 1px strokes *still* have 1px strokes, so text looks quite spindly!

    I was sure that I'd been told, by people here, that Windows 7 introduced revised ClearType, and wondered why ClearType still sucked on every Windows 7 machine I used. I've since learnt that only WPF gets that, so who knows how many years it will be before all Windows programs move to WPF and finally get smooth text. I guess never, because we'll have 1200 DPI displays long before then and the argument will become moot. Hell, at high DPI even regular antialiasing will be perfectly adequate, no need for ClearType in the first place, which is a win because ClearType is an awful hack really. (The fun phase will be screenshots taken from 300 DPI machines being shown on 100 DPI screens …)



  • @Shinhan said:

    Heh, when I had to restrict copy/paste in the internal app I made, I just counted keypresses. If the number of keypresses (backspace and delete decrement the counter ofc) is different from the length of the string, you get error.

     

    What about us Unix users with a compose key?  For example € is typed as [COMPOSE] [C] [=] (US keyboard so I don't have it as a key).  

     Hint on Linux: Go into your keybaord settings and change the [Menu] key to compose if you have a 104 key keyboard.  I am still useing a 101 key Model M so I use Right Control.



  • @ShawnD said:

    What about us Unix users with a compose key?

    You use Unix?

    @ShawnD said:

    I am still useing a 101 key Model M

    Natch.



  •  @blakeyrat said:

    You use Unix?

     Started with Solaris in university, now a heavy Linux user with some OpenBSD use in the past.  I guess Linux should be called Unixish.

     

     

    @blakeyrat said:

    @ShawnD said:
    I am still useing a 101 key Model M

    Natch.

    Found it in a trashcan years ago.  It is one of the ones with a built in trackpoint, but the ribbon cable to the buttons is broken and I haven't got around to trying to fix them yet.  Otherwise it has been perfect.

     



  • @ShawnD said:

    Started with Solaris in university, now a heavy Linux user with some OpenBSD use in the past.  I guess Linux should be called Unixish.

    (Knew it wasn't Unix. I win again.)

    @ShawnD said:

    Found it in a trashcan years ago.  It is one of the ones with a built in trackpoint, but the ribbon cable to the buttons is broken and I haven't got around to trying to fix them yet.  Otherwise it has been perfect.

    Ah, so your OS and hardware is trash.

    /troll



  • @ShawnD said:

    Found it in a trashcan years ago.  It is one of the ones with a built in trackpoint, but the ribbon cable to the buttons is broken and I haven't got around to trying to fix them yet.  Otherwise it has been perfect.

    You can still buy those new, although now only in the reduced fooprint model, Windows keys mandatory: Unicomp EnduraPro 104/105. Note that it will probably come with their garish logo, as seen on the plain SpaceSaver 104/105 – I paid them the extra $5 to customise it to have no logo. I went with the SpaceSaver as I never liked trackpoints anyway.

    The build quality isn't what IBM's was; it's not an aesthetic purchase (no black key option for example like the old double-shots), but you get a combination of buckling spring, 104/105 key layout, and USB that was never available from IBM.

    I live in hope that Unicomp will get some serious investment and refresh their very ageing moulds, slim up the cases to modern standards, and team up with one of the companies still able to produce double-shot keycaps. Model Ms are still in demand, but Unicomp's quality just isn't there. Still, their keyboards are pretty decent and feel great.



  • @blakeyrat said:

    @Shinhan said:

    Heh, when I had to restrict copy/paste in the internal app I made, I just counted keypresses. If the number of keypresses (backspace and delete decrement the counter ofc) is different from the length of the string, you get error.

    As for AltGr I use it all the time as well, all kinds of special characters I need are accessible for me only with AltGr: @{}<>[

    I always find it compelling when people tell these stories, with no apparent shame, of how they decreased the usability of their app.

    Yes because asking people to copy and paste a mistyped email address will really help when the emails start bouncing back.

    For many systems, email is the primary (or even sole) method of contacting your customers. When someone signs up and for whatever reason, mistypes their email address, you have LOST thatcontact method, and possibly the customer, for good.

    The purpose of entering your email again is to verify you typed it correctly ... if you just copypaste the first one, it has not been verified.

    Some things DO have a good reason behind them. Still, you stay "true" to your perceived user-friendly site, I'll make sure MY customers are contactable.



  • @daveime said:

    The purpose of entering your email again is to verify you typed it correctly ... if you just copypaste the first one, it has not been verified.

    Some things DO have a good reason behind them. Still, you stay "true" to your perceived user-friendly site, I'll make sure MY customers are contactable.

    Have you proven that requiring the user to enter the email twice helps? How did you test it, A/B test? Can I see the breakdown?

    Have you proven that the increased number of good contact emails you get (assuming there is an increase) outweighs the number of people who abandon your form after getting a "re-enter email" error? Do you have the form drop-off reports? Cost figures?

    You've come up with a theory, now where's the proof?


  • Discourse touched me in a no-no place

    @daveime said:

    For many systems, email is the primary (or even sole) method of contacting your customers. When someone signs up and for whatever reason, mistypes their email address, you have LOST thatcontact method, and possibly the customer, for good.

    Which is why you send a token in an email to that customer as part of the sign up process, asking them to verify their email, and don't open the account until that token has come back.



    This also has the possible advantages of

    • stopping people entering emails like fuckyou@example.com (with the expectation of getting an account with it.)
    • dissuading the creator of the site from using dodgy regexes which prevent entirely legitimate email addresses since the most they have to, prior to sending the token to begin with, is ensure there's an MX record for the bit after the @

    Some things DO have a good reason behind them. Still, you stay "true" to your perceived user-friendly site, I'll make sure MY customers are contactable.
    If all you're doing is making sure those two boxes have the same text in them, you're not making sure your customers are contactable.


  • This also has the possible advantages of...

    ...preventing people entering genuine emails other than their own, in order to embarrass/annoy/spam. I'll make sure my customers are who they say they are.


Log in to reply