|
Otaku Alpha Male posted:Flight computer crashes Excuse me stewardess, but I speak meme
|
#
?
Jul 11, 2016 17:56
|
|
|
|
| # ? May 15, 2024 02:09 |
|
|
this thread is also applicable to the vast majority of automotive engineers
|
|
#
?
Jul 11, 2016 17:58
|
|
|
Cat Face Joe posted:this thread is also applicable to the vast majority of automotive engineers i lust for industrial exemption death
|
|
#
?
Jul 11, 2016 18:13
|
|
|
H.P. Hovercraft posted:it's me i'm the firm unchanging project scope, hallmark of the engineering discipline traditional engineering projects change scope all the time, it's usually terrible (see: pentagon wars) just because we have software firms that can iterate reasonably successfully in the face of requirements changes doesn't make them not engineering, it means they're playing to the unique strength of the medium
|
|
#
?
Jul 11, 2016 18:33
|
|
|
i am an engineer
|
|
#
?
Jul 11, 2016 18:56
|
|
|
Bloody posted:i am an engineer wrong forum
|
|
#
?
Jul 11, 2016 18:58
|
|
|
computeronomy and computerology
|
|
#
?
Jul 11, 2016 19:12
|
|
|
Cocoa Crispies posted:traditional engineering projects change scope all the time, it's usually terrible (see: pentagon wars) let me go ahead and link this for you since it always comes up 1. The product is only as good as the plan for the product. At the on-board shuttle group, about one-third of the process of writing software happens before anyone writes a line of code. NASA and the Lockheed Martin group agree in the most minute detail about everything the new code is supposed to do — and they commit that understanding to paper, with the kind of specificity and precision usually found in blueprints. Nothing in the specs is changed without agreement and understanding from both sides. And no coder changes a single line of code without specs carefully outlining the change. Take the upgrade of the software to permit the shuttle to navigate with Global Positioning Satellites, a change that involves just 1.5% of the program, or 6,366 lines of code. The specs for that one change run 2,500 pages, a volume thicker than a phone book. The specs for the current program fill 30 volumes and run 40,000 pages. "Our requirements are almost pseudo-code," says William R. Pruett, who manages the software project for NASA. "They say, you must do exactly this, do it exactly this way, given this condition and this circumstance." This careful design process alone is enough to put the shuttle organization in a class by itself, says John Munson of the University of Idaho. Most organizations launch into even big projects without planning what the software must do in blueprint-like detail. So after coders have already started writing a program, the customer is busily changing its design. The result is chaotic, costly programming where code is constantly being changed and infected with errors, even as it is being designed. "Most people choose to spend their money at the wrong end of the process," says Munson. "In the modern software environment, 80% of the cost of the software is spent after the software is written the first time — they don't get it right the first time, so they spend time flogging it. In shuttle, they do it right the first time. And they don't change the software without changing the blueprint. That's why their software is so perfect."
|
|
#
?
Jul 11, 2016 19:23
|
|
|
and of course buildings never have dumbass customers change designs on them
|
|
#
?
Jul 11, 2016 19:24
|
|
|
H.P. Hovercraft posted:let me go ahead and link this for you since it always comes up I don't think I'd call full losses on 2 of 5 vehicles as 'perfect'
|
|
#
?
Jul 11, 2016 19:26
|
|
|
hobbesmaster posted:and of course buildings never have dumbass customers change designs on them they don't it's why we have contracts in the first place unless you're tolkien about architectural details which isn't really the same thing at all ofc there's always provisions for adding additional scope or w/e but that requires an addendum contract which means a lot more money and time
|
|
#
?
Jul 11, 2016 19:28
|
|
|
otoh, the software probably functioned just fine right until it was scattered across western texas
|
|
#
?
Jul 11, 2016 19:28
|
|
|
there's a lady where i work whose title is Technical Marketing Engineer
|
|
#
?
Jul 11, 2016 19:31
|
|
|
yeah that's an available option when you don't have to impress investors/execs before cash runs out
|
|
#
?
Jul 11, 2016 19:35
|
|
|
H.P. Hovercraft posted:they don't governments love their change orders
|
|
#
?
Jul 11, 2016 19:38
|
|
|
maniacdevnull posted:I don't think I'd call full losses on 2 of 5 vehicles as 'perfect'
|
|
#
?
Jul 11, 2016 19:43
|
|
|
H.P. Hovercraft posted:"Most people choose to spend their money at the wrong end of the process," says Munson. "In the modern software environment, 80% of the cost of the software is spent after the software is written the first time — they don't get it right the first time, so they spend time flogging it. In shuttle, they do it right the first time. And they don't change the software without changing the blueprint. That's why their software is so perfect." this presumes both requirements that can be completely documented and don't change, and an end to the process, which isn't necessarily true for most software look at facebook: yeah the original goal was so zuck could prove to rooney mara's character that he wasn't a bad person, but the goal and definition of success for the project has grown and changed in the last thirteen years in completely unforseeable ways so much of the process for it doesn't depend on prototypes or similar design artifacts because modern web software engineering focuses on making "idk ship it to some percent of users and see what breaks" strictly better and cheaper than it's ever been
|
|
#
?
Jul 11, 2016 20:02
|
|
|
facebook is an advertising company
|
|
#
?
Jul 11, 2016 20:15
|
|
|
H.P. Hovercraft posted:facebook is an advertising company as such they depend on showing ads to users, which means they have to meet ever-changing requirements to be interesting while being fast and reliable or they don't get paid
|
|
#
?
Jul 11, 2016 20:19
|
|
|
you're missing the point iterative work on a computer or designing advertising for the internet isn't engineering i understand that it's preferable to make that kinda work sound more meaningful but phrases like Cocoa Crispies posted:modern web software engineering are p lol to the people who design things that matter
|
|
#
?
Jul 11, 2016 20:28
|
|
|
Otaku Alpha Male posted:Flight computer crashes
|
|
#
?
Jul 11, 2016 20:35
|
|
|
H.P. Hovercraft posted:you're missing the point as a software engineer, people's second lives depend on the quality of my dick bridges and cock condos
|
|
#
?
Jul 11, 2016 20:43
|
|
|
i may not agree w/ the w-hat fetus cannon but i will die defending the goons' right to have one
|
|
#
?
Jul 11, 2016 20:45
|
|
|
Gazpacho posted:remind me how software was implicated in those failures ?? the Cluster failure was because assertions were turned off + integer overflow
|
|
#
?
Jul 11, 2016 21:20
|
|
|
Just FYI, that space shuttle software article was written in 1996. For context, Windows 95 had just been dragged kicking and screaming into the world. Lots of people were still running Windows 3.1. PC software was almost universally terrible. The shuttle software group was also insanely expensive IIRC. Like, $25 million/year kind of expensive. Literally no one can afford to spend that much and take that long to make small, incremental changes to software that isn't safety critical. Doc Block fucked around with this message at 22:29 on Jul 11, 2016 |
|
#
?
Jul 11, 2016 22:24
|
|
|
OldAlias posted:the Cluster failure was because assertions were turned off + integer overflow The flight software for the Arianne 5 rocket that failed was written by a different team, at a different company, in a different country, for a different space program. And they probably didn't follow the shuttle software group's extreme practices. They just reused the flight software from the Arianne 4 and didn't bother testing it on the Arianne 5 (even in simulation) before doing an actual launch. Neither of the space shuttle's failures were caused by its software.
|
|
#
?
Jul 11, 2016 22:28
|
|
|
Doc Block posted:That group was also insanely expensive IIRC. Like, $25 million/year kind of expensive. Literally no one can afford to spend that much and take that long to make small, incremental changes to software that isn't safety critical. lmao like maybe as a species we could afford maybe a hundred different pieces of software
|
|
#
?
Jul 11, 2016 22:29
|
|
|
i am a nerd, i will not go quietly into the night, i will rage, rage against pedantically trivial things like usage of the word engineer until my dying breath
|
|
#
?
Jul 11, 2016 23:17
|
|
|
again, commercial software devs do it differently from public contractors not because they're all dumb/evil but because they work for business ideologues who insist on things like shipping demos and cutting out QA and forcing code reuse. this isn't even a secret like half the problem in the therac-25 case was the ideology-soaked customer rep who kept reassuring hospitals that the equipment couldn't fail another example: when i worked at realnetworks in the 2000s i was on a project to develop the next release of their game download application. think steam but for cheaper games (in fact i suspect that i was there because valve hired away the previous dev team). every few months a new product manager would come in with a completely different vision but of course we had to reuse all of the code written on the previous manager's watch, no excuses, and that is exactly the antipattern that led to both the therac-25 and ariane V failures Gazpacho fucked around with this message at 23:28 on Jul 11, 2016 |
|
#
?
Jul 11, 2016 23:20
|
|
|
Gazpacho posted:another example: when i worked at realnetworks in the 2000s loll
|
|
#
?
Jul 11, 2016 23:28
|
|
|
i'll put my resume up against anyone's
|
|
#
?
Jul 11, 2016 23:42
|
|
|
here's my resume: buffering...
|
|
#
?
Jul 12, 2016 00:39
|
|
|
Gazpacho posted:i'll put my resume up against anyone's okay but be warned, mine doesn't say "RealNetworks" on it
|
|
#
?
Jul 12, 2016 00:40
|
|
|
Suspicious posted:here's my resume: easy but funny
|
|
#
?
Jul 12, 2016 00:48
|
|
|
Cocoa Crispies posted:okay but be warned, mine doesn't say "RealNetworks" on it Lol same
|
|
#
?
Jul 12, 2016 01:18
|
|
|
Cocoa Crispies posted:okay but be warned, mine doesn't say "RealNetworks" on it
|
|
#
?
Jul 12, 2016 01:18
|
|
|
oh sure i bet you just pick a listing you want and show up for work no questions asked
|
|
#
?
Jul 12, 2016 01:28
|
|
|
Cocoa Crispies posted:this presumes both requirements that can be completely documented and don't change, and an end to the process, which isn't necessarily true for most software The space shuttle was a safety critical system This is why verification and validation are so important in the prototyping stage You can't afford to fix it later because if one crashes it's a complete disaster Which is why Tesla's """autopilot"""" is an insult to control engineers
|
|
#
?
Jul 12, 2016 01:49
|
|
|
My posting qualifies under the shitpost cottage industry exemption
|
|
#
?
Jul 12, 2016 02:06
|
|
|
|
| # ? May 15, 2024 02:09 |
|
|
lol at pen and paper pushing fuckers trying to steal the the working man's credibility. railroad, marine, and operating engineers kick your asses.
|
|
#
?
Jul 12, 2016 02:16
|
|
Failed Sega Accessory Ahoy!