/dev/sigh, Where Developers Sigh

March 22, 2012 11:23 AM

/dev/sigh :: user-contributed scenes from the sometimes frustrating world of software and web development.
posted by milquetoast (43 comments total)

These make me feel better about (almost) every job I've ever worked at.
posted by tylerkaraszewski at 11:41 AM on March 22, 2012 [5 favorites]

Every developer has sighed for at least one of these things. At least one.
posted by mysterpigg at 11:46 AM on March 22, 2012 [2 favorites]

The "did you look at that problem during your free time?" bit kills me.
posted by Space Coyote at 11:50 AM on March 22, 2012 [3 favorites]

This is why I have a diploma that says "programmer" and a job title that says "support". At least I am expecting ignorance.
posted by Dark Messiah at 11:55 AM on March 22, 2012

client: why does 5 lines code take 2 hour
developer: coz I had to tested 10 other ways before finalizing these 5 lines as best possible approach
client: why does 5 lines code take 2 hour
developer: i hope you understand that LOC/hour is the worst possible metric for productivity
client: why does 5 lines code take 2 hour
developer: coding is the smallest part in the timelog, thinking and debugging takes the most time
client: why does 5 lines code take 2 hour

Once, in an offhand conversation, I convinced the software lead not to bow down to silly LOC numbers that management requested at their biweekly meetings. I kinda wish I was in the room to hear how management (a quite competent guy, just not a software guy) responded.
posted by mysterpigg at 11:59 AM on March 22, 2012 [2 favorites]

There also needs to be a /mgr/sigh and a /biz/sigh to handle the converse cases.

e.g.

"But the requirements only said to archive the data. Nobody mentioned anything about being able to restore it as well."

Sigh.
posted by philipy at 12:01 PM on March 22, 2012 [3 favorites]

Ugh, I read until the entry quoted below before I was immobilized under the dead weight of sympathetic despair that crashed down on my shoulders.

Friday at 4:30pm (before off at 5pm and off for weekend)

* CEO is back from meeting
<CEO> okay guys... let's push latest changes to live now.
<devs> sigh

posted by invitapriore at 12:02 PM on March 22, 2012 [11 favorites]

mysterpigg: "Every developer has sighed for at least one of these things. At least one."

Dude, you're being way too optimistic about this....
posted by schmod at 12:07 PM on March 22, 2012 [4 favorites]

"But the requirements only said to archive the data. Nobody mentioned anything about being able to restore it as well."

Well, if archiving is all you need, /dev/null does a great job with very low storage requirements.
posted by Philosopher Dirtbike at 12:16 PM on March 22, 2012 [7 favorites]

* CEO is back from meeting
okay guys... let's push latest changes to live now.
sigh

At my last job we specifically banned any release from going out on Friday except for emergency hotfixes for exactly this reason.
posted by tylerkaraszewski at 12:17 PM on March 22, 2012

client: why does 5 lines code take 2 hour
developer: coz I had to tested 10 other ways before finalizing these 5 lines as best possible approach
client: why does 5 lines code take 2 hour
developer: i hope you understand that LOC/hour is the worst possible metric for productivity
client: why does 5 lines code take 2 hour
developer: coding is the smallest part in the timelog, thinking and debugging takes the most time
client: why does 5 lines code take 2 hour

Once, in an offhand conversation, I convinced the software lead not to bow down to silly LOC numbers that management requested at their biweekly meetings. I kinda wish I was in the room to hear how management (a quite competent guy, just not a software guy) responded.

"It didn't. The 5 lines of code took two minutes. Discovering that those were the exact 5 lines of code that needed to be there, by researching methods and testing implementations and tracking down bugs, is what took two hours."

Don't accept the premise of the question!
posted by kafziel at 12:31 PM on March 22, 2012 [1 favorite]

client: why does 5 lines code take 2 hour

I've had a small but nonzero amount of luck getting through to people about why SLOC is stupid with the following phrase:

"You don't pay me to write code. You pay me to figure out what code to write."
posted by Tomorrowful at 12:32 PM on March 22, 2012 [20 favorites]

Client: We want to do this great Augmented Reality app for the iPad (first gen)
*sigh*
posted by ryoshu at 12:36 PM on March 22, 2012

wondering if should post about that one client that once requested (a+b)*0.5 to be changed to (a+b)/2 ..
posted by 3mendo at 12:44 PM on March 22, 2012 [1 favorite]

Sometimes frustrating?
posted by spamguy at 12:54 PM on March 22, 2012

"You don't pay me to write code. You pay me to figure out what code to write."

or...

There is an old story about a multi-million dollar power plant that had mysteriously ground to a halt. All efforts to restart it had failed and an expert was brought in. After studying the problem for a few minutes he took a hammer and hit one of the valves. With a rumble, the plant came back to life. Incredulous glances were shared, grateful cries and high-fives were exchanged. Later, the expert’s bill arrived for the amount of $10,000.00. The outraged executive in charge thought “All he did was hit a valve with a hammer, this bill is ridiculous.” he asked for an itemized breakdown and the consultant responded with a bill that read: “Hitting valve with hammer $10.00. Knowing which valve to hit: $9,990.00.”

posted by titus-g at 12:59 PM on March 22, 2012 [11 favorites]

"You don't pay me to write code. You pay me to figure out what code to write."

Reminds me of that old joke with the engineer and the $50,000 chalk mark:

There was an engineer who had an exceptional gift for fixing all things mechanical. After serving his company loyally for more than 30 years, he happily retired.

Several years later the company contacted him regarding a seemingly impossible problem they were having with one of their multi-million dollar machines. They had tried everything and everyone else to get the machine fixed, but to no avail. In desperation, they called on the retired engineer who had solved so many of their problems in the past.

The engineer reluctantly took the challenge. He spent a day studying the huge machine. At the end of the day, he marked a small "x" in chalk on a particular component of the machine and proudly stated, "This is where your problem is." The part was replaced and the machine worked perfectly again.

The company received a bill for $50,000 from the engineer for his service. They demanded an itemized accounting of his charges. The engineer responded briefly:

One chalk mark: $1
Knowing where to put it: $49,999

It was paid in full and the engineer retired again in peace.

posted by mysterpigg at 1:00 PM on March 22, 2012 [1 favorite]

Third-party jinx.
posted by kafziel at 1:04 PM on March 22, 2012

I'm really glad that most of these entries don't actually use:

*sigh*

I have a friend who uses that mannerism in almost every communication. I hate it.
posted by gurple at 1:11 PM on March 22, 2012 [1 favorite]

Ha!
posted by mysterpigg at 1:13 PM on March 22, 2012

You've been jinxed, motherfucker. Mail me some coke.
posted by kafziel at 1:14 PM on March 22, 2012

To manage people, shouldn't you at least either (a) know something about what they do or (b) trust them?
posted by maxwelton at 1:15 PM on March 22, 2012 [3 favorites]

I like this line: "haggling with me over reality." Ooh, that's so right. People who try to disagree with physics and space-time are half the problem in tech.
posted by Mo Nickels at 1:25 PM on March 22, 2012 [4 favorites]

My god I've been lucky...
posted by mmrtnt at 1:29 PM on March 22, 2012

I am not a coder, and this one made me want to scream into my pillow:

180
Anonymous sighs at:
The other intern at work: "I'm surprised anyone bothers with input validation."
ARRRRRRRRGH.

That, and the "work on your free time" one that Space Coyote mentioned.
posted by Xoebe at 1:31 PM on March 22, 2012

Get call at 4:30 am to support product I've never heard of. Find out I am the contact for every production product my entire division owns. Kick support ticket to network ops cuz fuck those guys.

Shit, I think I may be one of the bad guys.
posted by Ad hominem at 1:34 PM on March 22, 2012 [2 favorites]

Explaining why a fully-fledged website cannot be had for "like 200 bucks."
posted by inauthentic at 1:42 PM on March 22, 2012 [2 favorites]

pb should post some of the pony requests from Metafilter users.
posted by Philosopher Dirtbike at 1:53 PM on March 22, 2012 [2 favorites]

mysterpigg: "One chalk mark: $1
Knowing where to put it: $49,999"

If the accounting department had any humor, they'd send him a $49,999 check, and a note asking for a receipt for the chalk.
posted by schmod at 2:05 PM on March 22, 2012 [8 favorites]

Ooooh those bosses always doing silly things with SVN while trying to skimp on the dev's environment - and those pesky clients not gettin' shit and making crazy demands. Seems like nothing changed since the last time I saw a site like this.
posted by mattoxic at 2:29 PM on March 22, 2012

This one describes my job so very, very painfully and accurately.

manager : ok.. you can start doing task A now..
me : ok, I'm on it...
2 days later....
manager : hi.. I think you should doing this (task B) first, It is more urgent to finish by the end of this week.
4 days later task B finish, 2 days later
manager : why does it take so long to finish task A?

Or to put it another way: "Why does it take so long to do development work?" - "Because you have us all working on customer service issues."
posted by yath at 2:48 PM on March 22, 2012 [1 favorite]

pb should post some of the pony requests from Metafilter users.

Thinking about this (thanks, Philosopher Dirtbike), I'd like to see this as an annual event, a slow December end-of-year review thing, bracing the MeFi family for the new year.
posted by SPrintF at 4:24 PM on March 22, 2012

I've been encountering this kind of stuff in all kinds of fields and it makes me wonder how anything gets done at all, ever. It seems like every operation is just barely holding together with bits of string and tape and driftwood. I don't know how anything manages to get off the ground.
posted by bleep at 5:13 PM on March 22, 2012

Sigh, I've had all these issues, though I haven't had quite the level of psychopathy displayed in some of these stories.

One of my favorite moments was when I was sitting with four managers on a project that was behind - even though I'd been working ridiculous hours, making good progress and generating almost no bugs there were simply too many moving parts because of massive feature creep due to poor management.

All four of them were, fairly politely, explaining to me that I had to work faster. At one point I said, "You know, if there were one more of me and one less of you..."

They had to laugh.

At another time in the same company I was arguing with the bosses over a technical decision. I stopped and said, "You know, we've been having arguments like this for over a year, and every time you've been eventually forced to concede that I was right." There was a pause - they were honest men - and then one of them said, "Well, this time we're right." (In fact, it was a worse mistake than any other...)

I'm not sure how to avoid these sorts of problems, except to brutally interview the places you're thinking of working, never to work in a place which uses magic instead of reasoning, and to establish technical dominance over or at least parity with your colleagues early in your career there.

The last clause might seem a little aggressive, but you need to have enough standing that you can do things like admit ignorance, ask for an explanation of something a second or third time, or simply argue against an idea that you yourself brought up without raised eyebrows. In a functioning development community, these are easy - in these dysfunctional places, you can lose a lot of face to the point where no one will take you seriously.
posted by lupus_yonderboy at 5:16 PM on March 22, 2012 [3 favorites]

> I don't know how anything manages to get off the ground.

It turns out that access to capital is much more important than anything else, and putting together a company is so potentially lucrative that you can slap one together pretty badly and still make money with it - if you have access to capital.
posted by lupus_yonderboy at 5:17 PM on March 22, 2012

Well okay, getting a company up and running with capital is one thing, then how do they stay running. And also, why do companies even have employees if the managers and directors have no trust in or knowledge of their employees skills and take everything that's been done and decisions that have been made and just rip them up. I don't understand why you would hire anyone if you're just going to rip up everything they do.
posted by bleep at 5:46 PM on March 22, 2012

> then how do they stay running.

As I said, it's access to capital. The investors are buying a lottery ticket. The tech market is so lucrative that if one of these lottery tickets hits, it pays for all the others. The investors hire management and management hires the programmers. The investors can tell good ideas from bad ideas, but the investors are completely ignorant about technology, and also probably have multiple irons in the fire.

So these companies run through their investment capital and going back for more until the investors decide the ticket has expired and then they attempt to sell the intellectual property to other clueless companies.

I worked for a short time for Earthweb, which was exactly and precisely like that. I worked for another company like that for several years, and they are apparently still in business eight years later, with products that I've never heard of - I'd love to direct you to their website but discretion prohibits me - and the last I heard two or three years ago they had never had a single month that was profitable.
posted by lupus_yonderboy at 5:59 PM on March 22, 2012 [1 favorite]

I don't understand why you would hire anyone if you're just going to rip up everything they do.

Because if people like that put even as much thought into it as you did, with that sentence, they'd probably tender their resignation.
posted by Dark Messiah at 6:01 PM on March 22, 2012

"// SIGH: ..."

Should be a source comment annotation, similar to TODO or FIXME.
posted by clvrmnky at 7:36 PM on March 22, 2012 [3 favorites]

client: why does 5 lines code take 2 hour

My co-worker reported in yesterday's staff meeting that he took three days to find a bug in our testing infrastructure (which has hundreds of thousands of LOC in it) and fixed it with a one line change.

On the opposite side, I once found a library that something like 2400 lines of code but only 200 lines if you ran it through uniq. The coder had written one two hundred line function, copied it 11 times and changed one line in each. And then apparently committed it to the main-line without anyone reviewing it. He had a hell of a SLOC count though.
posted by octothorpe at 5:51 AM on March 23, 2012

tylerkaraszewski: At my last job we specifically banned any release from going out on Friday except for emergency hotfixes for exactly this reason.

What, you mean Read-Only Friday?
posted by wenestvedt at 6:44 AM on March 23, 2012 [1 favorite]

and fixed it with a one line change.

I've more than once fixed a badly broken program just by *deleting* a single line that didn't make any sense.

(Ok, I commented it out first, just in case.)
posted by philipy at 9:43 AM on March 23, 2012

My co-worker reported in yesterday's staff meeting that he took three days to find a bug in our testing infrastructure (which has hundreds of thousands of LOC in it) and fixed it with a one line change.

That's programming. Sometimes you don't produce any new code for a week because you spend that entire week tracking down a bug. Friend of mine did that recently, bug turned out to be in the Linux kernel itself.
posted by kafziel at 10:26 AM on March 23, 2012