• Harrow (unregistered)

    If the producing agent and the consuming agent agree on the format, the specification can go p!ss up a rope.

    -Harrow.

  • Mathias (unregistered)

    I'm an Easy Reader, and I didn't understand any of this.

  • Jazz (unregistered) in reply to htrjytiut
    htrjytiut:
    mainframe web dev:
    "WebSphere MQ". This story must be from around 2005.

    It was originally 'MQ Series' and now IBM brands it as just 'MQ'.

    http://www-01.ibm.com/software/integration/wmq/

    Out of 21 occurrences of "MQ", precisely one is not preceded by "WebSphere" (and it's the first Google hit for "ibm mq", so presumably it isn't some old page that wasn't updated for the alleged rebranding).

    God DAMMIT, you people. I thought I had repressed all of my memories of working with WebSphere. I wasted an entire summer in Washington, DC trying to make WebSphere MQ do something reasonable. Now I'm going to have nightmares about J2EE applications. Thanks.

    (Captcha: nobis -- no Business Information Systems, please.)

  • Gunslinger (unregistered) in reply to operagost
    operagost:
    The developers clearly expect the end users to do this:
    <equipment_to_load> <item_description>item 1</item_description> <some_other_field>0</some_other_field> </equipment_to_load> <equipment_to_load> <item_description>item 2</item_description> <some_other_field>101</some_other_field> </equipment_to_load> <equipment_to_load> <item_description>item 3</item_description> <some_other_field>247</some_other_field> </equipment_to_load>
    Anyone who doesn't see the needless inefficiency in that is not intelligent enough to understand why it doesn't match their documentation.

    No, they didn't intend that (they documented what they intended), but it's all they tested and all anyone had ever done up to this point.

  • Carsten (unregistered)

    We just had that last week, if your Contractor uses xml example data to implement it: Run if you can.

    Otherwise throw massive entities and other advanced xml compositions with CDATA fields containing the own payload as early as you can on them as mandatory test cases.

    XML without a standard parser that enforces scheme definition is always a WTF.

  • knowingthepain (unregistered) in reply to PolarityMan
    PolarityMan:
    Pretty sure MQ never touches the message payload.
    Right, it will probably be the transformation extender which touches the payload.
  • Paul Neumann (unregistered) in reply to knowingthepain
    knowingthepain:
    Right, it will probably be the transformation extender which touches the payload.
    Bad touch. BAD TOUCH! wait, no good touch.
  • swschrad (unregistered)

    so, you add a comment on the order of

    // SFM section: it isn't documented, but without this stuff, it doesn't work either. upline significantly unhelpful. basically duplicate everything with this parameter and you go home happy before midnight//

  • HowItWorks (unregistered) in reply to swschrad
    swschrad:
    so, you add a comment on the order of

    // SFM section: it isn't documented, but without this stuff, it doesn't work either. upline significantly unhelpful. basically duplicate everything with this parameter and you go home happy before midnight//

    Also: // submitted to DailyWTF, Featured Article 20130129

  • (cs) in reply to ip-guru
    ip-guru:
    Just because nobody complains dose not mean all parachutes are perfect.
    I gonna use that. Verbatim.
  • Someone (unregistered)

    Heh. XML issues are always the most annoying/amusing.

    Just recently I implemented a system to generate "XML" which was then loaded into an external system for HR processing. Having been bitten by this before I asked the external system to provide samples of what I should load. What I got back wasn't too bad except that it had no root node and hence was not real XML. I ended up generating the XML on my side with a stupid <STRIPTHISTAG> element as the root node, converting it to a string, stripping out the useless tag, and then saving it to disk.

    Of course that's the smallest of the WTF's of that project. I ended up complaining to one of their developers when I saw them and he admitted they had their own custom XML parser which somewhat explained the stupidity.

  • noname (unregistered) in reply to Cloy McTrure
    Cloy McTrure:
    noname:
    ip-guru:
    Just because nobody complains dose not mean all parachutes are perfect.
    Well, you usually do not have a chance to complain after that.
    [image]
    what joke? I think noone complain because they can't when they dead
  • Valency (unregistered)

    I am a lowly hobby programmer, so I've never worked on enterprise stuff. WTF are "loadouts"? I only know the term from military aviation, and I'm assuming that they aren't converting differing ordnance configurations.

  • (cs) in reply to Counter
    Feature Articles:
    It wasn’t valid, according to the schema
    <pedantic> It wasn't well-formed XML, according to the specification for XML itself. </pedantic>
    Feature Articles:
    Adam had one final question: how did he submit corrections to the documentation?
    <pedantic> how did he submit bug reports?. </pedantic>

    But the idea of a disillusioned developer passive aggressively telling them to fuck off by rewriting the documentation to fit the buggy behavior is amusing to me.

  • Doctor_of_ineptitude (unregistered) in reply to jay
    jay:
    1. receive complaint 2. is there a complaint about this problem already on file? 3. yes: duplicate, reject complaint 4. no: must be user's fault, reject complaint

    We are pleased to report that our complaint file remains empty. The system must be working perfectly.

    We also have a complaint system that remains empty.

    1. Receive the complaint.
    2. Explain how firearms from before 1898 do not require a license and ballistics can't be reliable.
    3. Show them Colt 1877 Bulldog Gattling gun.
    4. ???
    5. Profit
  • (cs) in reply to PedanticCurmudgeon
    PedanticCurmudgeon:
    snoofle:
    article:
    you’re the only one that’s ever had a problem with this
    So either nobody else has ever used it, or everyone else has figured out the same thing and has a work-around, and more interestingly, possibly an explanation. Perhaps find one of them and inquire?
    I think it's more along the lines of "We've never listened to any complaints".
    Or no-one else had ever been adventurous enough to have more than one item_info per document.
  • (cs)

    After reading all articles on the site I'm wondering why does so much bad code and ignorance exist in IT.

    Is it simply a flaw in human nature that affects most people and results in the WTF's we read here or are we decent programmers simply so uber smart that we can create decent code?

    And why do so many IT "specialists" insist that their way is best while the industry changes practically daily?

    If you can't evolve with the technology, you don't belong in IT.

  • Meep (unregistered) in reply to ObiWayneKenobi
    ObiWayneKenobi:
    You ever notice how the story always paints somebody as a self-serving asshole? In this case it's the WebSphere MQ Lead who just acts like a douchebag to somebody trying to do their job.

    But people are that way in real life. It's easier to be a douche and tell people to get fucked than to actually fix things.

  • Al (unregistered)

    Been here..

    This whole story sounds like a recent experience of mine. In this case it was an American software vendor in the energy industry. The software we use has web services that process XML documents but for one service any second node instance was ignored. The software vendor (happy in their signed contract) have a UK support team that like to give answers that imply that the system is perfect and the users are useless. When evidence was provided the reply was that no one else had complained. Later on it turned out that only one company in Europe was actually using the interface - us.

  • Michael D (unregistered)

    Reminds me of the time I was asked to pull vCard data from a client in order to populate their user accounts in our system. I tried a few open source vCard libraries but they didn't work. I finally cracked open the vCard specs and found what should have been obvious from the start - they were not sending me valid data. Naturally, I was the only person to ever have a problem with it so it must be an issue on my side. So I did what any self-respecting consultant would do... I created a custom vCard library from scratch and billed them for every second.

  • Bugs, Bugs Everywhere (unregistered) in reply to dgvid
    dgvid:
    My initial reaction to newly found bugs in code that I'd thought was well tested is often to think "You're the only one that's ever had a problem with this." Experience has taught me to keep my mouth shut and assume that the problem is real unless and until I've proven otherwise. I guess the WTF here is that the lead developer isn't very experienced.

    That's because if someone else had had a problem with it, it would have been fixed already.

  • Norman Diamond (unregistered) in reply to Al
    Al:
    This whole story sounds like a recent experience of mine. In this case it was an American software vendor in the energy industry. The software we use has web services that process XML documents but for one service any second node instance was ignored. The software vendor (happy in their signed contract) have a UK support team that like to give answers that imply that the system is perfect and the users are useless. When evidence was provided the reply was that no one else had complained. Later on it turned out that only one company in Europe was actually using the interface - us.
    That's what you get for offshoring your development work. Now you know better, right?
  • Chris (unregistered)

    Did anyone else notice the unicorns?

    (CAPTCHA: acsii. I typed it wrong so many times...)

  • Marbles (unregistered) in reply to Jochen
    Jochen:
    After reading all articles on the site I'm wondering why does so much bad code and ignorance exist in IT.

    Is it simply a flaw in human nature that affects most people and results in the WTF's we read here or are we decent programmers simply so uber smart that we can create decent code?

    A large portion of people who read this site are probably in the top 10% or so of programmers. We're interested in it, and want to see how NOT to do things (to help with knowing how we should do things) etc.

    The rest of the programmers in the world don't care about what they do, or if they did come to a site like this, they'd look at the CodeSODs and say 'OK, that looks like a good idea, I'll use that next time'.

    Really, I used to think myself an average C++ programmer, but so many times I come across total crap on 'programming websites', with people saying 'no, that's wrong, do it this (even more crap) way' and with no one seeing the 'obvious' bugs (uninitialised variables, buffer overflows, race conditions etc).

    So, I've decided I must be a fairly decent programmer and 90% of programmers should be stacking shelves instead.

  • (cs) in reply to Someone
    Someone:
    Of course that's the smallest of the WTF's of that project. I ended up complaining to one of their developers when I saw them and he admitted they had their own custom XML parser which somewhat explained the stupidity.
    I bet their "developers" are the same breed of people that complain when you want them to, I don't know, maybe write some actual code in an interview. But is so beyond them of course. Yeah, it's so beyond them that they'd waste my boss's money coding up a yet another broken XML parser. Maybe it should be a standard interview thing not only to code fizzbuzz, but also use an off-the-shelf XML parser of your choice, in a language of your choice, with a browser available to look things up of course. You fail that, you don't get the job. These days it's hard to avoid XML when dealing with almost any kind of software development, so IMHO it should be a hard requirement: you don't know how to use an existing one, you're not considered fit for the job.
  • Norman Diamond (unregistered) in reply to Marbles
    Marbles:
    So, I've decided I must be a fairly decent programmer and 90% of programmers should be stacking shelves instead.
    You know what would happen if 90% of programmers were stacking shelves? failblog would get more youtube videos.
  • yawn (unregistered)

    Can't be bothered to read these inane comments - it's a normal enterprise wtf, you don't know the context of the parent node - get off my lawn please

  • mainframe web dev (unregistered) in reply to htrjytiut

    Sorry for the confusion. The branding went MQ Series->WebSphere MQ->MQ.

    I suspect that next it will be either Rational MQ or possibly Jazz MQ.

  • Robert White (unregistered) in reply to whiznat

    I worked at a company that repeatedly failed to make an IPO because they could never get disclosures together that didn't show the company as totally boned up. Then they got a private buy-out offer. In the week before they finished the sale they ran some lay-offs that included all of the accountants.

    For years the company had been bidding government contracts at 70% of expected costs and then shifting money from new contracts to finish the old ones. The ultimate pyramid scheme.

    They laid off the accountants so that nobody who knew what was happening could spill the beans before the owners made off with the cash.

Leave a comment on “Scheming Schema”

Log In or post as a guest

Replying to comment #:

« Return to Article