• neminem (unregistered)

    I am also in agreement that a. users should have at least one, preferably multiple easy ways to report bugs in a system, but b. those channels should not be routed directly from users to the developers of the system. Rather, they should go through tech support first, tech support should verify that they're actually bugs and not already known issues, and then file a bug to the dev bug tracker, and the devs should get a notification to look into it at their convenience. This does, require, however, that tech support not be chained to a retarded script, and actually have both the intelligence and the privs to actually help customers, which seems to not be the case at most larger companies, where most of the job of tech support seems to be "get the user to hang up in disgust as quickly as possible, so we don't have to spend money on tech support."

  • (cs) in reply to neminem
    neminem:
    I am also in agreement that a. users should have at least one, preferably multiple easy ways to report bugs in a system, but b. those channels should not be routed *directly* from users to the developers of the system. Rather, they should go through tech support first, tech support should verify that they're actually bugs and not already known issues, and *then* file a bug to the dev bug tracker, and the devs should get a notification to look into it at their convenience. This does, require, however, that tech support not be chained to a retarded script, and actually have both the intelligence and the privs to actually help customers, which seems to not be the case at most larger companies, where most of the job of tech support seems to be "get the user to hang up in disgust as quickly as possible, so we don't have to spend money on tech support."

    If they had a really good grasp on the technical details of each task, they would instantly be promoted out of level 1 tech support.

  • Tye (unregistered)

    I'm not understanding why the number of records gets reset to zero at the start of each week.

    But yes in general programmers no longer think about optimization. There was a time when it was important (and not just because its embedded and doesn't have a file system). Hardware is cheap, throw more hardware at it. I see it in the programmers around me. Where C applications that used to run on a Pentium90 with 64 Mb of ram being replaced with Java applications with barely any function differences now need a dual core 2 GHz machine with 2 Gb of ram. And what do they resort to but "Raise the minimum requirements".

  • jay (unregistered) in reply to savar
    savar:
    The WTFing Article:
    Sort of the opposite of Einstein’s version of insanity

    DEAR INTERNET, THIS IS NOT THE DEFINITION OF INSANITY. Also, no publicly available source attributes this to Albert Einstein. Please stop.

    My definition of insanity is repeating the same quote for the 10,000 time, no matter how clever it might have been originally, and expecting people to be just as impressed at your cleverness, as they were by the first 9,999 people who said the same thing.

  • jay (unregistered) in reply to savar
    savar:
    faoileag:
    savar:
    The WTFing Article:
    Sort of the opposite of Einstein’s version of insanity

    Also, no publicly available source attributes this to Albert Einstein. Please stop.

    http://www.brainyquote.com/quotes/quotes/a/alberteins133991.html is not a publicly available source?

    Thanks for the pedantry. Here's a corrected version for your review:

    Also, no publicly available primary source attributes this to Albert Einstein.

    In place of "primary", you could also substitute "authoritative", "reliable", or "not cheap shitty fucking brainyquote" without loss of generality.

    What do you mean no authoritative or reliable source? It's quoted on thedailywtf.com! What more reliable source do you need?

  • jay (unregistered) in reply to neminem
    neminem:
    I am also in agreement that a. users should have at least one, preferably multiple easy ways to report bugs in a system, but b. those channels should not be routed *directly* from users to the developers of the system. Rather, they should go through tech support first, tech support should verify that they're actually bugs and not already known issues, and *then* file a bug to the dev bug tracker, and the devs should get a notification to look into it at their convenience. This does, require, however, that tech support not be chained to a retarded script, and actually have both the intelligence and the privs to actually help customers, which seems to not be the case at most larger companies, where most of the job of tech support seems to be "get the user to hang up in disgust as quickly as possible, so we don't have to spend money on tech support."

    Hee hee hee. He used the words "intelligence" and "tech support" in the same sentence!

  • jay (unregistered)

    But, wait, one thing still puzzles me here.

    1. The program took 30 times as long to display a screen on Fridays as it did on Mondays.

    2. For each "real" record read, the program was reading a dummy table 2 or 3 times.

    3. The programmer eliminated the unnecessary reads.

    Assuming #2 is essentially constant, the program would still take 30 times as long on Friday as on Monday. Okay, I guess the run times fall from 1 second and 30 seconds to 1/3 second and 10 seconds. But this change should have had nothing to do with the ratio of the times. And there's still something wrong with a program that does 30 times as much work on Friday as it takes to get the same result on Monday.

  • (cs) in reply to jay
    jay:
    there's still something wrong with a program that does 30 times as much work on Friday as it takes to get the same result on Monday.

    I figured it was something like a "Sales Closed This Week" report that grouped report results by weekday.

  • The Big Picture Thinker (unregistered) in reply to faoileag
    faoileag:
    I absolutely love the way the argument $date of a static function is sanitized before being inserted as a string into a SQL statement template, which is then executed without further ado ;-)

    Then again, I don't know PHP - maybe the SySQLStmt constructor has some Bobby Tables sniffer incorporated.

    No, there's not. And they should be using parametrized inputs (on prepared statements). For MySQL, PHP has a built-in library called "MySQLi" which makes it quite easy (albeit with certain limitations and quirkiness). This article just showcases the typical laziness and stupidity that comes with PHP "developers". And I do code in PHP sometimes (when I have to) so I know there is a lot of laziness and stupidity out there.

  • Anonymous. (unregistered) in reply to savar
    savar:
    Thanks for the pedantry. Here's a corrected version for your review:

    Also, no publicly available primary source attributes this to Albert Einstein.

    In place of "primary", you could also substitute "authoritative", "reliable", or "not cheap shitty fucking brainyquote" without loss of generality.

    There's no evidence that any version of this quote was known prior to 1980. The earliest known occurrence of the quote, in the form "insanity is repeating the same mistakes and expecting different results", is in a Narcotics Anonymous text dated 1981. An American novelist named Rita Mae Brown used it in the "standard form" in her novel in 1983.

  • (cs) in reply to The Big Picture Thinker
    The Big Picture Thinker:
    No, there's not. And they should be using parametrized inputs (on prepared statements). For MySQL, PHP has a built-in library called "MySQLi" which makes it quite easy (albeit with certain limitations and quirkiness). This article just showcases the typical laziness and stupidity that comes with PHP "developers". And I do code in PHP sometimes (when I have to) so I know there is a lot of laziness and stupidity out there.

    More powerful languages are just equivalent to more powerful guns that developers shoot themselves with.

  • neminem (unregistered) in reply to chubertdev
    chubertdev:
    If they had a really good grasp on the technical details of each task, they would instantly be promoted out of level 1 tech support.
    jay:
    Hee hee hee. He used the words "intelligence" and "tech support" in the same sentence!
    Note: I said that from personal experience. Our company, being decently-sized but not gargantuan, doesn't really have any "level 1" support in that sense at all. It has tech support, its tech support all (at least hypothetically) knows what they're doing. They don't necessarily know all the technical details of everything, but they are trained at actually problem-solving and at all our systems - the only part of the interaction I've heard that's obviously scripted is them finding out whether the client actually has a current license that has support for the product they're asking about.

    But this is also geared towards businesses, not individual users. For that it does make sense to have first-level drone support, but it also makes sense to always have some people who aren't first-level drone support around, to help out with issues that are not retarded, or that aren't in the script, and the first-level support should at least be trained to recognize "this is not in my script, I will give it to someone else". And by someone else, I don't mean "the person next to me, who will read the same script, and then repeat until the user gives up". (Though more commonly what it actually means is "I will promise the user that someone else will call back, then hang up and never tell anyone else to call them back."

  • AP² (unregistered) in reply to chubertdev
    chubertdev:
    The Big Picture Thinker:
    No, there's not. And they should be using parametrized inputs (on prepared statements). For MySQL, PHP has a built-in library called "MySQLi" which makes it quite easy (albeit with certain limitations and quirkiness). This article just showcases the typical laziness and stupidity that comes with PHP "developers". And I do code in PHP sometimes (when I have to) so I know there is a lot of laziness and stupidity out there.

    More powerful languages are just equivalent to more powerful guns that developers shoot themselves with.

    PHP isn't a less powerful gun, it's a gun that jams half the time and shoots backward the other half.

  • F (unregistered) in reply to Nyctef
    Nyctef:
    F:
    "...the entire function could be replaced by “return strtotime('monday this week')..."

    Not prior to PHP5.3.0. On dimanche the given function returns the preceding Monday, whereas in PHP5.2.0 and earlier 'monday this week' means tomorrow.

    See the "changelog" section of http://www.php.net/manual/en/function.strtotime.php

    And here I was thinking PHP had actually done something cool for once. . It should also really depend on the locale (or have an option to do so) to specify if a week starts on Sunday or Monday

    Perhaps I should have been more explicit.

    Prior to PHP5.3.0, "this week" meant the 7 days of which "today" is the middle. Since 5.3.0, it means the ISO week (Mon-Sun) containing today.

    Therefore "monday this week" is in the future if today is vendredi, samedi or dimanche and we are executing under older PHP.

  • F (unregistered) in reply to Anonymous.
    Anonymous.:
    savar:
    Thanks for the pedantry. Here's a corrected version for your review:

    Also, no publicly available primary source attributes this to Albert Einstein.

    In place of "primary", you could also substitute "authoritative", "reliable", or "not cheap shitty fucking brainyquote" without loss of generality.

    There's no evidence that any version of this quote was known prior to 1980. The earliest known occurrence of the quote, in the form "insanity is repeating the same mistakes and expecting different results", is in a Narcotics Anonymous text dated 1981. An American novelist named Rita Mae Brown used it in the "standard form" in her novel in 1983.

    It could still have originated with Einstein. After all, he'd only been dead for a quarter of a century. Someone of his calibre wouldn't let a little thing like that get in the way.

  • Almafuerte (unregistered) in reply to faoileag

    It's Serge in French, I think.

    Sergio is either Spanish, Italian or Portuguese.

  • Norman Diamond (unregistered) in reply to F
    F:
    Nyctef:
    F:
    "...the entire function could be replaced by “return strtotime('monday this week')..."

    Not prior to PHP5.3.0. On dimanche the given function returns the preceding Monday, whereas in PHP5.2.0 and earlier 'monday this week' means tomorrow.

    See the "changelog" section of http://www.php.net/manual/en/function.strtotime.php

    And here I was thinking PHP had actually done something cool for once. . It should also really depend on the locale (or have an option to do so) to specify if a week starts on Sunday or Monday
    Perhaps I should have been more explicit.

    Prior to PHP5.3.0, "this week" meant the 7 days of which "today" is the middle. Since 5.3.0, it means the ISO week (Mon-Sun) containing today.

    Therefore "monday this week" is in the future if today is vendredi, samedi or dimanche and we are executing under older PHP.

    Locale still wouldn't fix it. A company's own locale (their very local locale) might set a week to begin on Sunday, but if some of their projects serve customers whose weeks begin on Sundays and some of their projects serve customers whose weeks begin on Mondays they'll need something else (more local than their corporate local locale, but wormholed to each customer).

    Or they might be importing oil from a supplier whose weeks begin on Saturdays. Or they might have some branches whose weeks begin on Saturdays and some branches whose weeks begin on Sundays all in the same country.

  • Norman Diamond (unregistered) in reply to F
    F:
    Anonymous.:
    savar:
    Also, no publicly available primary source attributes this to Albert Einstein.
    There's no evidence that any version of this quote was known prior to 1980.
    It could still have originated with Einstein. After all, he'd only been dead for a quarter of a century. Someone of his calibre wouldn't let a little thing like that get in the way.
    Insanity is arguing over and over again on Usenet and expecting the argument to stop this time.

    Hitler knew how to stop an argument on the internet.

  • WyuZenZuZex (unregistered)

    I'm fairly certain that the definition of insanity originated with Erickson. It's discussed at some length in "The Structure of Magic" the NLP book that John Bandler and Richard Grinder wrote about Erickson's work.

    Or something like that. It may have been one of the other books they wrote about Erickson.

  • Jeremy (unregistered) in reply to Tye
    Tye:
    Where C applications that used to run on a Pentium90 with 64 Mb of ram being replaced with Java applications with barely any function differences now need a dual core 2 GHz machine with 2 Gb of ram. And what do they resort to but "Raise the minimum requirements".

    Back in the day, programs ran fine on a $500 PC.

    These days, the modern counterparts to those programs still run fine on a $500 PC.

    I'm all for efficient coding, but spending tens of thousands of dollars in developer time optimizing a program to run on a PC that can no longer even be purchased is a waste of capital.

  • (cs) in reply to jay
    jay:
    But, wait, one thing still puzzles me here.
    1. The program took 30 times as long to display a screen on Fridays as it did on Mondays.

    2. For each "real" record read, the program was reading a dummy table 2 or 3 times.

    3. The programmer eliminated the unnecessary reads.

    Assuming #2 is essentially constant, the program would still take 30 times as long on Friday as on Monday.

    Most likely there was one initial query which returned the whole recordset. So on Monday we have one meaningful query and (say) 200 useless queries, on Friday we have one meaningful query (with more results) and 6000 useless queries. It's quite possible that the dominant factor in both cases was the useless queries.

  • faoileag (unregistered) in reply to Norman Diamond
    Norman Diamond:
    F:
    Anonymous.:
    savar:
    Also, no publicly available primary source attributes this to Albert Einstein.
    There's no evidence that any version of this quote was known prior to 1980.
    It could still have originated with Einstein. After all, he'd only been dead for a quarter of a century. Someone of his calibre wouldn't let a little thing like that get in the way.
    Insanity is arguing over and over again on Usenet and expecting the argument to stop this time.

    Hitler knew how to stop an argument on the internet.

    Dear Sir, I do have the vague feeling that you invoke Goodwin's Rule on purpose here, and I have to object: if invoked on purpose, Goodwin's Rule does not apply!

  • Anonymous (unregistered)

    TRWTF is letting users dictate software design or relying on them to define problems.

    It took years before we ever adopted a required "testing" phase of software (outside of the original developer), but now I think that the majority of my colleagues consider testing it to be running the application and clicking on the buttons. If nothing explodes and nothing looks amiss then it's good for production!

    Nevermind that we usually don't understand how it's supposed to work in detail (of course there are no formal specs of any kind) and serious and obvious bugs that should easily be spotted by peer review are completely ignored.

    "It works" is not good enough for me. I hope it never is.

  • (cs)

    "Looks like someone's got a 'case of the mondays'"...

  • Todd Lewis (unregistered) in reply to Enzo
    Enzo:
    After all, who the hell knows French?

    Right. Face it: English is the "Lingua Franca du Jour."

  • (cs) in reply to Norman Diamond
    Norman Diamond:
    Hitler knew how to stop an argument on the internet.
    I am fairly certain that it was not Hitler but actually a project called "Manhattan" that developed acurate means to stop an argument on the internet.
  • Norman Diamond (unregistered) in reply to no laughing matter
    no laughing matter:
    Norman Diamond:
    Hitler knew how to stop an argument on the internet.
    I am fairly certain that it was not Hitler but actually a project called "Manhattan" that developed acurate means to stop an argument on the internet.
    Actually no. A few years ago we had to report to a foreign government on our capability to continue operating even if one more metropolitan area in this country gets nuked. Since we have operations in both Tokyo and Osaka, if only one of them gets nuked the other can continue operating. Even when Tepco and the government collaborated[*] to let Fukushima get nuked, and Tokyo's electricity supply was subjected to rolling blackouts, Osaka had no problem. So the Manhattan project isn't enough to stop an argument on the internet.

    [* When Tepco and the government couldn't imagine the possibility of a tsunami being as big as the tsunami in 1896 and 1933, the government approved Tepco's inadequate safety measures.]

  • Norman Diamond (unregistered)

    Yes, just like humans, computers need time to recover from Wednesday Thursday Friday.

  • Dan (unregistered) in reply to Nagesh

    Dilbert said it best: it's the curse of competency

    (September 13, 2008, for the interested)

  • ¯\(°_o)/¯ I DUNNO LOL (unregistered) in reply to F
    F:
    "...the entire function could be replaced by “return strtotime('monday this week')..."

    Not prior to PHP5.3.0. On dimanche the given function returns the preceding Monday, whereas in PHP5.2.0 and earlier 'monday this week' means tomorrow.

    See the "changelog" section of http://www.php.net/manual/en/function.strtotime.php

    I'm surprised they didn't name the new version "real_strtotime".

    So will the next PHP support strtotime('lundi this week')? That would be a better match for this bilingual code.

  • (cs) in reply to Norman Diamond
    Norman Diamond:
    Actually no. A few years ago we had to report to a foreign government on our capability to continue operating even if one more metropolitan area in this country gets nuked. Since we have operations in both Tokyo and Osaka, if only one of them gets nuked the other can continue operating. Even when Tepco and the government collaborated[*] to let Fukushima get nuked, and Tokyo's electricity supply was subjected to rolling blackouts, Osaka had no problem. So the Manhattan project isn't enough to stop an argument on the internet.
    As a reminder - in that same country: 1.) Hiroshima 2.) Nagasaki
  • JJ (unregistered) in reply to faoileag
    faoileag:
    Dear Sir, I do have the vague feeling that you invoke Goodwin's Rule on purpose here, and I have to object: if invoked on purpose, Goodwin's Rule does not apply!
    Goodwin's Rule? Is that like the Goodrich Blimp?
  • a (unregistered) in reply to Anonymous Paranoiac

    [quote user="Anonymous Paranoiac"][quote user="Lis"][quote user="QJo"] Given them something that requires fuzzy decision making and they'll immediately ask for an automation.[/quote]

    That is because thinking is hard.

  • a (unregistered) in reply to a

    [quote user="a"][quote user="Anonymous Paranoiac"][quote user="Lis"][quote user="QJo"] Given them something that requires fuzzy decision making and they'll immediately ask for an automation.[/quote]

    That is because thinking is hard. [/quote]

    guess TRWTF is me in not counting the quote tags. Proof that Thinking is hard?

  • Matt (unregistered)

    I wonder what Monday's date is when today's $date is "') OR DROP TABLES--"

  • Robert (unregistered)

    Monday's date? Pfah! That's what week numbers are for.

  • Nameless (unregistered)

    I had a similar problem. I was stuck with optimization of one routine that took 4 hours but was required to complete in two. As it turned out that every time the date was needed, a function was called that did "SELECT sysdate FROM dual" and output the result to a variable. It took only 3ms (on average) to run this, but this was called so many times (a lot of nested loops) that using one variable that was populated at the beginning reduced running time more than an hour.

    Other optimizations were less WTFy.

Leave a comment on “Computers Need Weekends Too”

Log In or post as a guest

Replying to comment #:

« Return to Article