• Richard Marshall (unregistered)

    My toddler sibling watches Numberblocks. Lucky him he still can't read, or else he would start shouting at the developers, he likes correcting people in this matter vigorously. Not that I don't get him, I would do the same.

  • SUC 2 (unregistered)

    Numberblocks machete order

  • (nodebb)

    I can't recall any meetings that caused me to forget my location, but certainly I've often forgotten why I was there and, in extreme cases, what the point was in carrying on living.

  • (nodebb)

    Oh, one of my pet peeves: dialog buttons that don't match the question (canceling the cancel). Back when I was doing some UI programming I insisted on labeling dialog buttons such that they directly answered the question being asked. I always got push back from product management and always lost the battle. So much for trying to increase the level of intelligence in the universe.

  • tbo (unregistered) in reply to RogerC

    Isn't that a standard Javascript confirm() call in Chrome? The buttons are OK and Cancel; you don't get to change them. They'd have to change the question, I suppose. "Are you sure you want to discard your comment?"

  • Randal L. Schwartz (google)

    For the last one, I suspect the data in question is referenced by historical orders, and there's no provision for migrating those orders to hold a legacy address. And the programmer hadn't anticipated that (or wasn't paid enough to handle it), and it ended up in the "catch all -- report error to user even though they can't fix it" message handler.

  • (nodebb) in reply to RogerC

    Don't forget the "I shook my iPhone with a text entry field active" thing, where it asks you if you want to undo what you typed, with options "Undo" and "Cancel". But I really want to talk about the old version of that in French, where the buttons were "Annuler" meaning "Undo" and "Annuler" meaning "Cancel". Made it hard to know which was what.

  • Klimax (unregistered) in reply to Randal L. Schwartz

    Don't think so. This error is coming from MS-SQL when action tries to violate one of constraints like Foreign Index. What's likely happening, addresses are linked to orders. (Interestingly, it's not common in CZ/SK eshops to have even option to store more than two address - one main/billing and one shipping; if one needs to ship elsewhere they need to fill it manually each time) That's why it is common design to have normalization here.

    Not sure how to handle it. Either offer deactivation/hiding of old addresses or blanking it out, but one would have to check relevant laws. (Data retention)

  • (nodebb) in reply to SUC 2

    Oh my goodness, I'd never heard of machete order before. Thank you. For anyone else currently unaware, it's about the best order in which to watch the six main Star Wars movies. Check out https://www.rodhilton.com/2011/11/11/the-star-wars-saga-suggested-viewing-order/

  • israr (unregistered)

    [just for people following along and kibbitzing about comment spam, you might find this interesting. The following is the comment spam I'm about to delete, but I edited instead for you - lws] Oh my goodness, I'd never heard of a machete order before. Don't think so. For the last one, I suspect the data in question is referenced by historical orders, and there's no provision for migrating those orders to hold a legacy address. check out [url mangled for safety and anti-spam] apkofdark.blogspot.com ... [ obfuscated]...apk-latest-version-v19312.html [ I'm assuming this is probably malware, you're welcome ;) ]

Leave a comment on “Foreign Keys”

Log In or post as a guest

Replying to comment #:

« Return to Article