Pick your own, but don't go to the source


  • Trolleybus Mechanic

    It's currently prime time for pick-your-own berries in Ontario. It might be a nice outing, and hey, Google reminds me that there's a Pick Your Own webpage that'll let me know which farms have which berries. Go check it out.

    I'll wait.

    http://www.pickyourown.org/canadaon.htm

     Are you back? How're your eyes? I bet you think that's what TRWTF is, right? I mean, Comic Sans, multiple colors, seventy pages long, bizzare copyright notices... 

    Nope.  You want to know what TRWTF is?  Go ahead and view source.

    I'll wait.

    Oh, I should have asked-- are you using Firefox? Because if so, welcome back from your browser crash! Yes, folks, TRWTF is that View Source can crash Firefox. I wonder what it is in the source that can crash the browser. If you want to know, go to the page and view the source.

    I'll wait.



  • @Lorne Kates said:

    I wonder what it is in the source that can crash the browser. If you want to know, go to the page and view the source.

    I'll wait.

    Nothing. Literally. Lines 1793 through 132867 seem to have absolutely nothing in them.



  • Well, it almost crashed chromium,I closed out before it did, but I was able to look at the source by saving the page. It is 133218 lines long of which, if I counted right, about 131075 of those lines is a block of newlines and spaces. So about 98.4% of the source is whitespace.



  • <LINK rel=stylesheet type=text/css href="newtemplates/styleprint.css" media=print>

    Nice... capitalized tags? Check. Missing quotes around attributes? Check (mostly). Unclosed tags? Check. Freaking MEDIA-CONDITIONAL STYLESHEETS? CHECK...

    It's like the past 20 years of web development, all rolled into one!



  • What's the problem with media conditional stylesheets?



  • @ekolis said:

    Freaking MEDIA-CONDITIONAL STYLESHEETS? CHECK...

    Sorry, that fad is back in style again.



  • @Mcoder said:

    What's the problem with media conditional stylesheets?

    Nothing. I'm saying they're surprisingly modern compared to the rest of the site!



  • xmlns:m = "http://schemas.microsoft.com/office/2004/12/omml" xmlns:v = "urn:schemas-microsoft-com:vml" xmlns:o = "urn:schemas-microsoft-com:office:office"



    Wait, was it generated on MS Office?



    MSHTML 8.00.7600.16625



    When I search for it all I find are reference for other pages with this generator tag on the code.



  • @Lorne Kates said:

    It's currently prime time for pick-your-own berries in Ontario. It might be a nice outing, and hey, Google reminds me that there's a Pick Your Own webpage that'll let me know which farms have which berries. Go check it out.

    I'll wait.

     Are you back? How're your eyes? I bet you think that's what TRWTF is, right? I mean, Comic Sans, multiple colors, seventy pages long, bizzare copyright notices... 

    Nope.  You want to know what TRWTF is?  Go ahead and view source.

    I'll wait.

    Oh, I should have asked-- are you using Firefox? Because if so, welcome back from your browser crash! Yes, folks, TRWTF is that View Source can crash Firefox. I wonder what it is in the source that can crash the browser. If you want to know, go to the page and view the source.

    I'll wait.

    Chrome also crashed AFTER trying to load the source. For added fun I have to try this site on my mobile; if the entire north american 4G network collapses in the next few minutes don't be surprised.

    By the way, is it because your browser is traumatized that you could not put a proper link? Lazy complainers are the worse. *sigh* *rolling eyes* *more sigh*.


  • Trolleybus Mechanic

    @Speakerphone Dude said:

    For added fun I have to try this site on my mobile; if the entire north american 4G network collapses in the next few minutes don't be surprised.
     

    On an unrelated note, North American productivity went up for a few minutes today.

    @Speakerphone Dude said:

    By the way, is it because your browser is traumatized that you could not put a proper link? Lazy complainers are the worse. *sigh* *rolling eyes* *more sigh*.

    ... the fuck? I did put the url tag there. Goddamn CS.



  • Didn't crash for me.

    According to firebug, the page is 1022 KB. There were several calls to nonexistent resources. And yes lines 1792 - 132866 appears to be nothing but whitespace. It actually has two head sections! And one of them is contained within a div...

    Wait, just found a third head section. I'm surprised the site works at all.



  • @Mcoder said:

    What's the problem with media conditional stylesheets?

    Nothing.  What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?

     



  • @da Doctah said:

    @Mcoder said:

    What's the problem with media conditional stylesheets?

    Nothing.  What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?

     

    1. Community Server is horrible (given)
    2. Community Server = CS
    3. CS = Comic Sans
    4. Comic Sans is hoorible



  •  I gave up on waiting and killed the process.  Dunno whether it would've eventually crashed on its own.

     



  • @mott555 said:

    I'm surprised the site works at all.
     

    Here you see what's wrong with HTML and The Web in general. If you write any (say) Java or C# program and you write "pirvate" instead of "private", you're not even getting past compiling. With HTML, you can just put any random shit in there and it will most likely still work.

     



  • @da Doctah said:

    @Mcoder said:

    What's the problem with media conditional stylesheets?

    Nothing.  What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?

     

     

    What isn't the problem with Comic Sans?

     



  • @pbean said:

    Here you see what's wrong with HTML and The Web in general. If you write any (say) Java or C# program and you write "pirvate" instead of "private", you're not even getting past compiling. With HTML, you can just put any random shit in there and it will most likely still work.
     

    Well, yes, because HTML is a markup language, not a programming language. I'd love browsers to be less lenient, granted, but the fact that it'll always do SOMETHING (be it not what the designer intended maybe) is pretty much by design.

     



  • Specifically, I'd like it if a browser failed on malformed HTML.

    That's mostly it.



  • @Lorne Kates said:

    Go ahead and view source.

    I'll wait.

    Oh, I should have asked-- are you using Firefox? Because if so, welcome back from your browser crash!

    I just tried it in Firefox 11 under OS X 10.7, and it doesn't crash. Takes a while to display anything in the source window, but the browser keeps running; tried it in Firefox 3 for good measure, and no crash there either.



  • @da Doctah said:

    Nothing.  What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?
    Comic Cans lacks credibility.

    (that article is written in Georgia, so I moderately agree)



  • In Chrome, view source freezes the tab, and after wgetting it, it still crashes Notepad++. Visual Studio loads it okay, but freezes for a few seconds while doing so.

    The file itself contains 131,071 (according to Visual Studio's search-replace function) lines containing nothing save a tab and two spaces.

    I think we finally found the elusive "view source protection".



  • @MiffTheFox said:

    I think we finally found the elusive "view source protection".

    Works fine in less.

    Though, I've worked with log files over 800Gb with less, so it's not too surprising :P



  • @Lorne Kates said:

    TRWTF is that View Source can crash Firefox

     

     I cannot reproduce said crash. Yes, it takes firefox a minute or 2 on 100% CPU, but it does not crash, and eventually shows source..

     If any, this proves the urgent need for multi-core support when viewing source under ff.

     


  • Discourse touched me in a no-no place

    @nanobit said:

    urgent need
    You wot? @nanobit said:
    multi-core support when viewing source
    I wasn't under the impression that viewing source was an inherently parallelisable task.


  • ♿ (Parody)

    @PJH said:

    @nanobit said:
    multi-core support when viewing source

    I wasn't under the impression that viewing source was an inherently parallelisable task.

    Well, I was still able to scroll around in the page while the source was loading, so it seems to be parallelized at least from normal browsing operations. So either it's running in its own thread in that tab's process or is its own process. I didn't care enough to figure out which it was.



  •  FFx's viewsource basically loads the page again (form cache, probably, or live) and renders it as text instead of feeding it into gecko.

    You can even F5 it independently from the page it's "from".



  • @dhromed said:

    Specifically, I'd like it if a browser failed on malformed HTML.

    That's mostly it.

    You follow your dream and make that browser! You can call it "Dhromed's Browser That Doesn't Fucking Work". See how much uptake you get with your miracle browser which is great in every way except can't display 75% of the pages on the Internet.

    BTW, the W3C is firmly in your camp, which is why they spent so many years on the pointless and useless XHTML2 spec before finally being convinced by some miracle that it was actually pointless and useless and switched to HTML5.



  • The idea is good but you can't do it because the hole is already too deep.

    Just to be sure: I'm only referring to closing tags and attribute quotes. Everything else about strictness can go fuck itself.

     



  • Trolleybus Mechanic

    @blakeyrat said:

    You can call it "Dhromed's Browser That Doesn't Fucking Work"
     

    If you're going to make something shitty that doesn't work, I think it's law that it has a recursive acronym*.

    Dhromed's Browser "DBDDFW" Doesn't Fucking Work.

     * I might have cause and effect reversed.



  • @dhromed said:

    The idea is good but you can't do it because the hole is already too deep.

    So the idea is bad. Got it.



  • @blakeyrat said:

    @dhromed said:
    The idea is good but you can't do it because the hole is already too deep.

    So the idea is bad. Got it.

     

    Lots of ideas are awesome if you ignore reality for a moment.



  • @dhromed said:

    @blakeyrat said:

    @dhromed said:
    The idea is good but you can't do it because the hole is already too deep.

    So the idea is bad. Got it.

     

    Lots of ideas are awesome if you ignore reality for a moment.

    Communism and bacon-wrapped helicopters spring to mind.



  • @Someone You Know said:

    Communism and bacon-wrapped helicopters spring to mind.
     

    Economy! Religion!

     

    ... agile

    ... coding patterns



  • @dhromed said:

    Economy coding patterns! agile Religion!


    FTFY



  • @Zecc said:

    @da Doctah said:
    Nothing.  What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?
    Comic Cans lacks credibility.
    It has more credibility in my eyes than any font in which lowercase ell is indistinguishable from capital eye or number one.

    Which makes it better than Arial, Courier or Times New Roman.



  • @dhromed said:

    Specifically, I'd like it if a browser failed on malformed HTML.

    That's mostly it.

     

    You are luck, there are several versions of Firefox out there that'll fail on malformed HTML. (By failing you meant like segfault, didn't you? Or did you meant refuses to display, but keeps working? If it is the last one, don't mind, I missinterpreted you.)

    @da Doctah said:

    What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?

    Nothing, except that everybody did independently arbitrarily decide the same "I don't like it".

     



  • @serguey123 said:

    @dhromed said:

    Economy coding patterns! agile Religion!


    FTFY

    "Oh noes! The rapture hasn't occurred yet! Guess we'll have to push back our armageddon deadline by another thousand years!"



  • @pbean said:

    What isn't the problem with Comic Sans?
     

    Comic Sans does have one redeeming feature: it is one of the easiest-to-read commonly available fonts for many people with dyslexia.

    But aside from that, your post is brilliant.

     



  • Maybe you should go to the genius creator's website and tell me what you think... Better yet, why don't you click on the search button in his website and tell me what you think... Better yet why don't you peruse through THE FUCKING ROOT DIRECTORY OF HIS WEBSITE and tell me how many files you can run/read when you shouldn't be able to (this is my favorite directory by the way, the table file basically gives away his entire db schema).

    Only people who are, not even in the slightest, tech savvy, such as farmers, would hire this guy for a job. I also like how his slogan is "Innovation Made Practical". 



  • @Mcoder said:

    @da Doctah said:

    What's the problem (apart from "I've arbitrarily decided that I don't like it") with Comic Sans?

    Nothing, except that everybody did independently arbitrarily decide the same "I don't like it".

    For suitably small values of "everybody", and excluding those who instead decided "I once heard <influential person> say that Comic Sans...was bad, m'kay?"

     



  • @captainpants said:

    Maybe you should go to the genius creator's website and tell me what you think... Better yet, why don't you click on the search button in his website and tell me what you think... Better yet why don't you peruse through THE FUCKING ROOT DIRECTORY OF HIS WEBSITE and tell me how many files you can run/read when you shouldn't be able to (this is my favorite directory by the way, the table file basically gives away his entire db schema).

    Only people who are, not even in the slightest, tech savvy, such as farmers, would hire this guy for a job. I also like how his slogan is "Innovation Made Practical". 

    Thank you for this link. Now I've made the mistake to start looking at their customers websites, such as the beautiful Consumer Fraud Report

    ... and I've identified a possible COLLUSION SCHEME involving their customers: look at the pickyourown, pumpkins and Christmas tree sites and notice the link where one can go to add his farm: it's the SAME!!! That blake2009@pickyourown.org is a clever businessman.

    EDIT: name resolution for all the customers websites comes from NS1.PICKYOUROWN.ORG... including benivia.com! The only possible explanation is that PickYourOwn has been such a commercial success that they took over the competition as well as the consulting firm that created their website.



  • @PJH said:

    @nanobit said:
    urgent need
    You wot? @nanobit said:
    multi-core support when viewing source
    I wasn't under the impression that viewing source was an inherently parallelisable task.

    Oh you poor old OOP of course you can run in parallel this kind of shit:

    T1: Load First two pages of data (I don't know, 200KB?) and render one page to screen
    T2: Load more pages into memory from those 200KB
    T3: Read the whole fucking file so it knows how much data you'll need in the future and can page it easily.


  • BINNED

    With both FF and Chrome (almost) crashing I was quite surprised Opera didn't complain. So out of curiosity I started poking around.

    W3C Validator:

    Sorry, I am unable to validate this document because on line 275 it contained one or more bytes that I cannot interpret as utf-8 (in other words, the bytes found are not valid values in the specified Character Encoding). Please check both the content of the file and the character encoding indication. 
    

    The error was: utf8 "\xA0" does not map to Unicode

    Oh, nice. Ok, time to try tidy:

    Info: Document content looks like HTML Proprietary
    764 warnings, 4 errors were found!
    

    This document has errors that must be fixed before
    using HTML Tidy to generate a tidied up version.

    Character codes 128 to 159 (U+0080 to U+009F) are not allowed in HTML;
    even if they were, they would likely be unprintable control characters.
    Tidy assumed you wanted to refer to a character with the same byte value in the
    specified encoding and replaced that reference with the Unicode equivalent.

    Choked. Ok, I have the line number, let's do it by hand:

    <option value="info.htm">How*much*to*I*need*to*pick?</option>
    

    NOTE: offending chars replaced by *

    In an option tag too! Brillant!

    Trying to run through W3C validator now results in clipboard not even entertaining the notion of pasting that mess. Trying to upload the file results in

    Proxy Error
    

    The proxy server received an invalid response from an upstream server.
    The proxy server could not handle the request POST /check.

    Reason: Error reading from remote server

    Apache/2 Server at validator.w3.org Port 80

    Are my clipboard and W3Cs site broken or just too scared? You decide!



  • @Onyx said:

    Oh, nice. Ok, time to try tidy:
    Try -raw parameter (which you want anyway when you're processing anything with tidy, because otherwise it'll quietly convert all smart quotes to regular plain boring normal ones).@Onyx said:
    Are my clipboard and W3Cs site broken or just too scared? You decide!
    Try p.0au.de or sprunge.us.


  • BINNED

    @ender said:

    @Onyx said:
    Oh, nice. Ok, time to try tidy:
    Try -raw parameter (which you want anyway when you're processing anything with tidy, because otherwise it'll quietly convert all smart quotes to regular plain boring normal ones).@Onyx said:
    Are my clipboard and W3Cs site broken or just too scared? You decide!
    Try p.0au.de or sprunge.us.

    I really don't care much for the specific case but I did save those links for future reference.

    Also, today was the first time I ever even used tidy. Because, you know, I tend to keep my own code tidy anyway. But I appreciate the heads-up ($deity knows I'll have to deal with someone else's BS sooner or later, my luck is bound to run out). Also, it still choked after I cleaned up the non-unicode mess. I can't blame it.



  • Amazing.  IE 7 has absolutely no problems loading the source of that page.  Of course, since it just dumps the source into notepad and makes no attempt to parse it, it should not be much of a surprise.



  •  @dhromed said:

    Just to be sure: I'm only referring to closing tags and attribute quotes. Everything else about strictness can go fuck itself.

    Should expand that to include things like having multiples of things that a page should only have one of (e.g. multiple heads or bodies) and cross-nesting (e.g. <em><strong></em></strong>*).

    I would also include interpretting all JavaScript errors as fatal, but that would break every site ever. Shit, I'd be willing to bet there's a site out there that uses setTimeout as the JavaScript version of On Error Resume Next.

     

    * I once saw a browser - I think Opera - actually accept this, and do what you'd guess. I would assume it automatically broke the strong tag into two parts rather than writing some unholy hackery to actually make a sensible DOM tree out of this (a DOM Ewok village?). Which brings to mind the old software engineering principle: When in Rome, do as Microsoft Word does.

     



  • @curtmack said:

    * I once saw a browser - I think Opera - actually accept this, and do what you'd guess. I would assume it automatically broke the strong tag into two parts rather than writing some unholy hackery to actually make a sensible DOM tree out of this (a DOM Ewok village?).
    Inspecting this sentence with Firefox's developer tools and Opera's dragonfly reveals that they both do the same thing - produce a sensible DOM tree from the cross-nested tags.



  • @ender said:

    @curtmack said:
    * I once saw a browser - I think Opera - actually accept this, and do what you'd guess. I would assume it automatically broke the strong tag into two parts rather than writing some unholy hackery to actually make a sensible DOM tree out of this (a DOM Ewok village?).
    Inspecting this sentence with Firefox's developer tools and Opera's dragonfly reveals that they both do the same thing - produce a sensible DOM tree from the cross-nested tags.

    Oooh, good call. For those who don't feel like opening up dev tools, the DOM tree looks like this:

    <i>
        this
        <b>
            sentence
        </b>
    </i>
    <b>
        with
    </b>

    That's from the View HTML option in Firefox developer tools, which shows how Gecko actually built the DOM tree (and is updated if any page content is altered by a script); View Source just shows the verbatim source it received from the HTTP GET. (Incidentally, it shows the invalid nesting as an error.)

    I imagine this approach makes some amount of sense for formatting tags, but I'd hate to see what it produces when you start doing this with, say, a div and a script.

    Edit: Just checked. Firefox will make no attempt to cleanly reconcile the following source:

    <html>
    <head>
    <title>WTF?!</title>
    </head>
    <body>
    <div>
    I'm a div!
    <span>
    And I'm a span!
    </div>
    Die div, die!
    </span>
    </body>
    </html>

    Output DOM (focusing only on the div/span clusterfuck):

    <div>
        I'm a div!
        <span>
            And I'm a span!
        </span>
    </div>
    Die div, die!

    It seems that when closing an outer block, it automatically closes all inner blocks, unless those blocks can be split in two without changing their function (e.g. font, b, i), in which case it will do that instead. It will always produce a warning in View Source about the incorrect nesting though.


Log in to reply