Forgot your password?
typodupeerror
Bug Technology

2016 Bug Hits Text Messages, Payment Processing 340

Posted by Soulskill
from the y2k16-has-more-characters-than-2016 dept.
An anonymous reader writes "It seems some systems are suffering from a Y2K16 bug. When 2009 ticked over to 2010, some Australian EFTPOS machines skipped to the year 2016. Coincidentally, some Windows Mobile users are also having issues with their new year SMSes coming from 2016. What function could cause this kind of error?"
This discussion has been archived. No new comments can be posted.

2016 Bug Hits Text Messages, Payment Processing

Comments Filter:
  • Why (Score:5, Funny)

    by Anonymous Coward on Sunday January 03, 2010 @09:04AM (#30631388)

    Uncle Bill decided to do some coding. That's why.

    • Re: (Score:3, Interesting)

      by TheLink (130905)
      Nah, it's more likely the same bunch that did excel 2007... Anyone remember the infamous 77.1 * 850 = 100000 Excel 2007 bug? :)
      • by Anonymous Coward

        It is 2016, not 2k16. Is there something cool about replacing a zero with a k now?
         
        Any EE will tell you that 2k16 means 2.16k or 2160. How does this garbage continue making it to the front page?

        • by Chabil Ha' (875116) on Sunday January 03, 2010 @10:54AM (#30631986)

          No to mention Y2K was a pun of sorts for shortening 2000 to three bytes from four. Now we're taking more space than the original; Y2K16 vs 2016. Those COBOL programmers would be rolling in their graves.

        • Re: (Score:3, Insightful)

          by osu-neko (2604)

          Any EE will tell you that 2k16 means 2.16k or 2160. How does this garbage continue making it to the front page?

          Actually, most EE's I know are smarter than that. The fact that X means Y in some contexts does not mean X means Y in all contexts. Most competent speakers of the language have no trouble determining meaning, taking context into account. It's rare that someone manages to get an EE while lacking the level of intelligence required to do that.

  • Some kind of... (Score:5, Insightful)

    by msauve (701917) on Sunday January 03, 2010 @09:06AM (#30631394)
    BCD/binary mismatch?

    0x09 = 9 decimal when interpreted as either binary or BCD.

    0x10 = 10 decimal when interpreted as BCD, as 16 when interpreted as binary.
    • Guess that makes the most sense. Although I have a hard time imagining how to do this ... but considering how complex the code probably is, and in what hurry that Y2K fix was probably done, it's the most likely scenario.

    • Re: (Score:3, Interesting)

      by mrslacker (1122161)

      Well, certainly BCD confusion caused a similar problem for RISC OS. Machines have been skipping to 2012. More here:

      http://www.riscository.com/ [riscository.com]

  • 4-bit years? (Score:2, Insightful)

    by Anonymous Coward

    A 4-bit year field? Wrapping from 9 to 0 (16) ?

  • Microsoft (Score:5, Informative)

    by sopssa (1498795) * <sopssa@email.com> on Sunday January 03, 2010 @09:08AM (#30631402) Journal

    Seems Microsoft is supplying some code [microsoft.com] for EFTPOS machines that is common with Windows Mobile, so it's most likely the same bug in both.

  • by Gorobei (127755) on Sunday January 03, 2010 @09:10AM (#30631410)

    No dial tone, no incoming calls.

    Had to reset the internal datetimes back to 2007, then they started working again. Nice job, Panasonic.

  • by nurhussein (864532) on Sunday January 03, 2010 @09:11AM (#30631414) Homepage
    Imagine that, the ability to text to the past. We could get warnings about impending events! New Text Message From Marty McFly: I HV TO TELL U ABT TEH FUTURE Reply from Doc Brown: WHAT? New Text Message From Marty McFly ON THE NIGHT YOU GO BACK, YOU GET SH*message text missing*
    • and they can bill you the future rate as well. In the Year 2016 texts are $2 each way. and your bill in 6 years past due.

  • 10 hex is 16 decimal (Score:5, Informative)

    by Anonymous Coward on Sunday January 03, 2010 @09:11AM (#30631416)

    Could be botched string parsing. Could be binary coded decimals interpreted as binary numbers: BCD encodes two decimal digits in the high and low nibbles of a byte. Therefore BCD 10 is 0001 0000 in binary, which is 16 in decimal.

    • by nycguy (892403) on Sunday January 03, 2010 @09:20AM (#30631456)
      This is almost certainly what it is. The year is stored in an SMS message as a two-digit BCD value, according to this spec [dreamfabric.com]. (Click on the link [dreamfabric.com] for the "timestamp" field.) Some phones must be treating it as a hex field. (Note that most other fields in the SMS message are in hex.)
      • I thought the whole Y2K fiasco taught people not to use two digits to represent years. Am I the only one who learned that lesson?
        • by Skapare (16644)

          Two digits is fine, now ... until we approach 2100. I plan to be retired by then, so I don't care. Fear those who are using ONE digit (that has happened). But in this case I can believe that someone used the wrong conversion like maybe: strtol(year_str,NULL,0x10).

          • Two digits is fine, now

            Apparently not.
          • These messages had better not be spending years getting routed around the phone company network. The phone can infer what year it is.

            You don't need a month even. I'm not sure you need a day, and you could probably skip the rest as well. This is a stupid waste of bytes, and thus air-time. It eats away at space that could go to the message or anything else useful.

            Suppose we really do want sender-side time. OK, let's use an 8-bit count of minutes. That covers over 4 hours. We take UNIX time (UTC seconds since

            • by pipatron (966506)
              Too bad if your recipient turned off the phone during the night, ran out of batteries during a hike, or changed number temporary due to a trip abroad (roaming costs fucking sucks).
          • Two digits is fine, now ... until we approach 2100.

            Wrong. It's only fine until we approach 2010. "Microsoft... nothing is impossible!"

        • I thought the whole Y2K fiasco taught people not to use two digits to represent years.

          Correct. However, you’re ignoring a significant piece of data that can be used to reconstruct the full year from only the last two digits.

          Context.

          Your cellphone already knows the current full year, and nobody will keep records of SMS messages for 100 years (and if they did, it would be in a summary form that showed the full year of its creation).

          In other words, use the first two digits of the current year (which you know), and if that creates a date in the future, subtract 100. The only time this meth

        • by Pembers (250842) on Sunday January 03, 2010 @10:32AM (#30631862) Homepage

          The specification for the SMS message format pre-dates Y2K by about 15 years. I came across it in 1995 (and thought it was useless - what sort of idiot would try to fit a message into 160 characters? And who would want to type it on a 12-button keyboard?). Where I worked, nobody worried about Y2K until about 1998.

      • by Posting=!Working (197779) on Sunday January 03, 2010 @10:20AM (#30631798)

        Can someone tell me what the advantage of swapped nibble encoding are? Other than just being annoying as fuck when you're trying to decode it?
        For those too lazy to read the link , swapped nibble encoding is
        "BCD code where nibbles within octet is swapped. E.g.: 0x31 Represents value of 13"

        So for the format YY MM DD HH MM SS TZ (Time zone in 15 minute increments from GMT) instead of 10 01 03 10 11 43 24 for 2010 Jan 3 10:11:43 time zone 24 (GMT +6) you get the identical data but in the less readable form of 01 10 30 10 11 34 42 (and now it can be confused for 2001 Oct 30 10:11:34 AM. Bonus!)

        It's just complete idiocy to me. Is there some reason you'd want the date/time stamp slightly harder to read?

        • Re: (Score:2, Insightful)

          by Anonymous Coward

          That may be an attempt to distinguish the BCD values from normal binary numbers. The only value for which swapped nibble encoding equals binary encoding is 0, so a test-case would catch a misinterpretation easily. I've seen a unit test which tests the algorithms with 2005 for the year. That test will not catch an error in a single-byte-year algorithm which mistakes non-swapped BCD for binary or vice versa.

        • Re: (Score:3, Informative)

          by clone53421 (1310749)

          It makes a lot more sense in little endian.

        • by Anonymous Coward on Sunday January 03, 2010 @01:33PM (#30633196)

          The endianness may make it more sensible. If you have 0x12345678 represented using the correct endianess (the Intel one, not the Sun one), it'll be stored as 0x78563412. Using reverse nibble notation, it'll be 0x87654321.

          But as far as I remember, it is caused by oldold ties to the 4004, which was a 4-bit CPU, inherited via the 8008 and 8080 by the 8086 and 8088.

      • by Megane (129182)

        They're doing YEAR = BCDYEAR + 2000... hooray for clueless programmers!

        What makes it stupider is that with SMS they have to swap the nibbles first... and decode the other fields as BCD too.

        And this is clearly a client-only bug, so dozens of codebases will have to be fixed and millions of devices updated. The typical lifetime of a cellphone means that the broken ones may simply get trashed rather than go through the trouble of an update.

  • by kju (327) *

    It's a Y2K10-Bug. The Y2K-Bug was not titled Y1900-Bug (to name one of the bug types) as well.

    • Y2K10 would be 200010. Silly contractions...

      • by TeknoHog (164938)

        Y2K10 would be 200010. Silly contractions...

        Then again, in electronics 2K10 would be parsed as 2.10 thousand, or 2100.

    • I was actually considering that something like this might happen, due to the Y2K craze. People are scrambling to "fix" that hyped problem, testing being done haphazardly and only to make sure that it works "for now" because we have "more time later".

      Then later, nobody cared about it anymore and nobody cared about the next decade. I'm actually quite happy that the biggest problem seems to be that text messages come from the future. Hey, I didn't care about those coming from the past either, so I'm fine with

  • by Anonymous Coward on Sunday January 03, 2010 @09:18AM (#30631448)

    It's to avoid the world ending in 2012 by skipping straight to 2016. We've left a few years either side of the fateful date as a safety buffer.

  • by rescendent (870007)
    The beam collisions have opened a portal into the future...
    • Huh, I always thought the LHC would cause a resonance cascade.

      • I always thought the LHC would cause a resonance cascade.

        Yes, but a temporal resonance cascade. It can send people forward to 2016, or backwards to 1620, depending on your cell phone's software.
  • Start worrying about SMSs that warns about future events. The "avoid seeing that movie" sms could have been because a limited preview somewhere, but im still a bit worried about the "the alien landing on Jan 8 will be real" sms.
  • by amn108 (1231606) on Sunday January 03, 2010 @10:00AM (#30631684)

    Time to quote Joel (from www.joelonsoftware.com):

    Jamie Zawinski is what I would call a duct-tape programmer. And I say that with a great deal of respect. He is the kind of programmer who is hard at work building the future, and making useful things so that people can do stuff. He is the guy you want on your team building go-carts, because he has two favorite tools: duct tape and WD-40. And he will wield them elegantly even as your go-cart is careening down the hill at a mile a minute. This will happen while other programmers are still at the starting line arguing over whether to use titanium or some kind of space-age composite material that Boeing is using in the 787 Dreamliner.

    When you are done, you might have a messy go-cart, but it’ll sure as hell fly.

    Hey, Jamie! Your proprietary date datatype didn't fly! Would you please turn around and fix it?

    • by daveime (1253762) on Sunday January 03, 2010 @11:16AM (#30632094)

      Yes, but that's just the point isn't it ?

      There are 10 types of people ...

      1. Those who will make a solution that will work for perhaps 30 or 35 years, because it is "good enough" for the foreseeable future, ala Y2K, 2032 for linux etc ...

      2. Those who will want to do unit tests to see if the date function still works in the year 9500, and won't pass it through QA until it does.

      Now, consider the PHB, when you tell him that solution 1 will be up and running in 5 minutes, whereas solution 2 will be up and running in about 5 years, once the specification has been formalized and ratified by the UN security council.

      Which solution will the PHB choose I wonder ? Programming has always been a compromise, no one *really* expects that something they code today will still be around in 35 or 40 years, it's only occasionally that something *does* survive that long (COBOL legacy systems etc), and come and bite us in the ass when the new millenia arrives.

      Having said that, I wonder how many of the "buzzword" languages we see today will still be around in the next 40 years ? The concept of longevity might have been relevant in the 60's, these days it's more a case of a complete rewrite every year or so.

      • by fnj (64210)

        Dude, it's 2038 for 32 bit Unix/linux, not 2032. 64 bit is good until some time after the Sun's end of life.

        The guys who designed in the 2038 limit not long after 1970 made a solution that would last 68 years. I call that pretty good. Long enough to be damn sure to be retired or dead before trouble hits.

    • by Blakey Rat (99501)

      In Jamie's defense, I've read that book, and Joel's interpretation of what he said is completely, totally, 100% off-base. Plus the name "duct-tape programmer" is offensive. Usually, I'm a fan of Joel. But then again, usually he's not so wrong and offensive in his articles.

      (The sad thing is that I agree with the general principle of the article, but the way it's presented, and the way it's attributed to Jamie, is completely off-base.)

      • by amn108 (1231606)

        I personally think Joel is one damn smart fellow. After reading his biography, I am also pretty sure he is a very dependable and likeable human being.

        The original comment was indeed a joke. I do think duct tape programming has its uses, despite what I earlier argumented for (the topic was slashdotted some time ago), and I do think Jamie by now is probably earning his PhD. It is just that I do also think the type of error the 2016 bug is can indeed be attributed to the "duct tape programming" attitude. I've

        • Re: (Score:3, Informative)

          by Blakey Rat (99501)

          It is just that I do also think the type of error the 2016 bug is can indeed be attributed to the "duct tape programming" attitude.

          But that's exactly what Jamie *isn't* talking about. He never said anything about not designing the system, or doing the design ad-hoc, all he's saying is two things:

          1) If your code is too complex, uses too many obscure language features, or has a structure that can't easily fit in the average person's head, it's likely to be more buggy since some of your team will never fully u

  • by olsmeister (1488789) on Sunday January 03, 2010 @10:01AM (#30631688)
    Bob- Happy Nw Yr! What a decade so far. Cubs winning the series, and who would have predicted that the Mayan calendar was really right? Glad I loaded up on all the MSFT back in '10. Are you going to visit your wife's grave next week for the 6th anniversary? kthx!
    • You sir, are an evil, evil person. Also, your ideas are intriguing to me and I wish to subscribe to your newsletter.

    • who would have predicted that the Mayan calendar was really right? Glad I loaded up on all the MSFT back in '10.

      Although I agree that MSFT performing well would be quite a (figurative) catastrophe, but in case of a real catastrophe, what good does it do to load up on any kind of stock certificate?

      And yes, stock certificates are generally printed on paper that's far too stiff and not absorbant enough for the "obvious" purpose...

  • by ub3r n3u7r4l1st (1388939) * on Sunday January 03, 2010 @11:49AM (#30632302)

    This is an indication that the human race will survive 2012! Rejoice!

  • no that's a Y210 bug (Score:4, Informative)

    by rubycodez (864176) on Sunday January 03, 2010 @11:58AM (#30632366)

    Y bugs are named for the year in which they occur, not the year they jump....otherwise Y2K would have been Y1.9K, or even better YMCM

The ideal voice for radio may be defined as showing no substance, no sex, no owner, and a message of importance for every housewife. -- Harry V. Wade

Working...