|
I don't know what DFSS is, but from context it seems to be some kind of time machine?
|
# ? Nov 8, 2019 11:12 |
|
|
# ? Apr 28, 2024 13:24 |
|
Teriyaki Hairpiece posted:I thought it was "hooses". No that's just any group of houses in Scotland.
|
# ? Nov 8, 2019 11:26 |
|
Goon Danton posted:I don't know what DFSS is, but from context it seems to be some kind of time machine? It's a business process that uses regression analysis to (supposedly) optimize a manufacturing or corporate process. I briefly had to work with it in the original dot-com era, and it's a total clusterfuck. It's mostly a way for upper management to justify their decisions with meaningless correlations, also they get to fetishize the Japanese. I'm mildly horrified that anyone is still taking it seriously.
|
# ? Nov 8, 2019 20:25 |
|
Ohhhh it's six sigma! Statistics but stupid and also a cult! But what's the DF? Dwarf Fortress Six Sigma? Are we making process control charts for our plump helmet harvests?
|
# ? Nov 8, 2019 20:33 |
|
Goon Danton posted:Ohhhh it's six sigma! Statistics but stupid and also a cult! But what's the DF? Dwarf Fortress Six Sigma? Are we making process control charts for our plump helmet harvests? DF stands for daddy fetish.
|
# ? Nov 8, 2019 20:37 |
|
Paladinus posted:DF stands for daddy fetish.
|
# ? Nov 8, 2019 20:38 |
|
Housii
|
# ? Nov 8, 2019 20:39 |
|
Right now, in some boardroom someone is showing a PowerPoint slide with the stonks meme as the conclusion. The cocaine kicks in and the board begins to celebrate.
|
# ? Nov 8, 2019 21:35 |
|
Goon Danton posted:Ohhhh it's six sigma! Statistics but stupid and also a cult! But what's the DF? Dwarf Fortress Six Sigma? Are we making process control charts for our plump helmet harvests? My guess is "Design For."
|
# ? Nov 8, 2019 21:49 |
|
Tobermory posted:It's a business process that uses regression analysis to (supposedly) optimize a manufacturing or corporate process. I briefly had to work with it in the original dot-com era, and it's a total clusterfuck. It's mostly a way for upper management to justify their decisions with meaningless correlations, also they get to fetishize the Japanese. I'm mildly horrified that anyone is still taking it seriously. people still believe in NPS too; humans love their measurements, even if they don't mean anything
|
# ? Nov 8, 2019 22:00 |
|
Tobermory posted:It's a business process that uses regression analysis to (supposedly) optimize a manufacturing or corporate process. I briefly had to work with it in the original dot-com era, and it's a total clusterfuck. It's mostly a way for upper management to justify their decisions with meaningless correlations, also they get to fetishize the Japanese. I'm mildly horrified that anyone is still taking it seriously. Just to be clear, none of that should be confused with statistical capability for manufacturing processes, where the term six sigma comes from originallly. That's pretty much the only way to actually make large production runs of parts, where measuring every part is time-prohibitive. The actual statistics people use are kinda BS, someone threw together some equations without too much statistical background. But they work well enough when treated as a black box, which is how everyone does it these days because nobody understands them. It's good enough. Everything else is utter bullshit. Managers and consultants took lean manufacturing and made the simple idea of "apply common sense to improve value and remove waste" as complicated as possible in order to justify their existence. Then they stole language from actual manufacturing processes in order to make it seem legitimate. Karia has a new favorite as of 22:25 on Nov 8, 2019 |
# ? Nov 8, 2019 22:14 |
|
Let's take this method designed for analysing defects on a production line churning out a million identical items, and apply it to software development! What could possibly go wrong?
|
# ? Nov 8, 2019 22:45 |
|
I've only ever dealt with six sigma people in a manufacturing scenario and it was kinda Goon Danton has a new favorite as of 00:57 on Nov 9, 2019 |
# ? Nov 8, 2019 23:27 |
|
Goon Danton posted:I've only ever dealt with six sigma people in a manufacturing scenario and it was kinda fun even then. People are trying to use it for writing code? How? MBAs are a hell of a drug
|
# ? Nov 8, 2019 23:49 |
|
Goon Danton posted:I've only ever dealt with six sigma people in a manufacturing scenario and it was kinda fun even then. People are trying to use it for writing code? How? It's easy! https://www.isixsigma.com/tools-templates/software/six-sigma-meets-software-development/ iSixSigma posted:Six Sigma Versus SDLC To translate: we figure out what the customer needs (problem definition/stakeholder analysis.) We figure out the relevant variables (discovering data relationships) and flowchart it (process mapping.) Then we solve the problem. Then we measure the results to make sure that the problem's actually been solved, not just papered over (measuring for quality.) If you think that sounds like what every software engineer does anyway, then you aren't an MBA! Unfortunately, though, there are a lot of industries and companies where even that incredibly basic idea of data-driven decision making and approaches really haven't taken hold too much. That's where Six Sigma and other, similar systems come in. SS is a bunch of simple tools and basic analysis flowcharts that consultants can come in to tell management to implement, independent of the industry. Often, they can find bottlenecks or waste in the process and fix them. I remember one case study of a naval shipyard where they shaved months off of construction time by changing their paperwork system. In short, they actually can have some pretty impressive results! The thing is, though, that you could get results just as good by going and talking to the engineers who are signing all of the papers and asking them what the holdup is. Six Sigma is a top down system. The goal is to highlight problems so management can see them. Then the consultants can suggest solutions and management can feel like they helped the situation (even though they were the ones who probably caused the problems in the first place by demanding lower downtime, not providing money for repairs, requiring levels of bureaucracy, etc.) Let's contrast with the Toyota Production System, one of the original and most famous lean manufacturing implementations. One of the most notable things about TPS is that everyone at the company, from management down to the assemblers putting bolts on cars, has the responsibility to find problems and suggest solutions. The famous example here is that anybody on the production line can pull a cord or push a button and stop the entire production line if they see an issue with a car. The goal is to get everyone at the company involved in continuous improvement. TPS does have standardized approaches, too, but it's generally less structured, and relies more on common sense than on blind application of standard formulas or tools. Toyota also had a culture where the managers all came from the actual operations: they knew cars and knew manufacturing, so everyone spoke the same language and they could build a system that worked for them. A few years back, I remember hearing that Toyota was having issues with that system. There was new management who were putting a ton of pressure on people not to stop the line, and they were having a lot more quality issues. Good management is invisible: its job is to facilitate the people who actually do the work. Bad management gets in the way and ruins everything. Thus concludes more information than any sane person (read: non-MBA) will ever want to know about Six Sigma. EDIT: Oh, to be clear as well: Six Sigma the management principle doesn't even really use statistical capability like is in its name, which is a manufacturing concept. Six Sigma has nothing to do with six sigma. So it's not like they're trying to calculate Ppk for the number of lines of code you write versus the number that have errors or anything. When I say Six Sigma "tools", what I'm referring to is stuff like this. It's something obvious (identify the ultimate rather than proximal cause for a problem) that's been turned into a buzzword piece of dogma that can be applied without understanding why that's important. Karia has a new favorite as of 00:55 on Nov 9, 2019 |
# ? Nov 9, 2019 00:34 |
|
Goon Danton posted:I've only ever dealt with six sigma people in a manufacturing scenario and it was kinda fun even then. People are trying to use it for writing code? How? People have been desperately trying to find ways to measure software production with hard numbers as long as software production has been a thing. Software developers are really loving expensive so they want to wring as much money out of them as possible. The snag is that this isn't, say, widgets rolling off of a production line and you're also measuring the output of tremendous nerds with math skills that are going to find a way to optimize whatever number you measure as the best to the detriment of everybody else so good loving luck. It's even more problematic when the decisions are made by people who have never written a line of code in their lives and can't be bothered to ask the code monkeys because, hey, how hard even is programming, anyway?
|
# ? Nov 9, 2019 00:35 |
|
Somfin posted:MBAs are a hell of a drug I once had an apparently fresh MBA regional manager talk about workers on the factory floor. We were a bookstore chain.
|
# ? Nov 9, 2019 02:20 |
|
Workin' at the sales factory.
|
# ? Nov 9, 2019 02:23 |
|
ToxicSlurpee posted:People have been desperately trying to find ways to measure software production with hard numbers as long as software production has been a thing. Software developers are really loving expensive so they want to wring as much money out of them as possible. The snag is that this isn't, say, widgets rolling off of a production line and you're also measuring the output of tremendous nerds with math skills that are going to find a way to optimize whatever number you measure as the best to the detriment of everybody else so good loving luck. It's even more problematic when the decisions are made by people who have never written a line of code in their lives and can't be bothered to ask the code monkeys because, hey, how hard even is programming, anyway? The double hilarious thing is that programmers are wildly expensive, but the hyper rational people employing them usually refuse to hire dedicated tech writers and project managers, who are cheaper and better at what they do than programmers
|
# ? Nov 9, 2019 02:35 |
|
CMS posted:The double hilarious thing is that programmers are wildly expensive, but the hyper rational people employing them usually refuse to hire dedicated tech writers and project managers, who are cheaper and better at what they do than programmers Testing is another vital element that managers skimp on in many ways.
|
# ? Nov 9, 2019 04:03 |
|
It basically boils down to the reason programmers are so expensive is because companies would rather pay them 150% to do three jobs than pay three people at 100%, and then get all pissy when it turns out you get bad code out of that process.
|
# ? Nov 9, 2019 04:40 |
|
The Cheshire Cat posted:It basically boils down to the reason programmers are so expensive is because companies would rather pay them 150% to do three jobs than pay three people at 100%, and then get all pissy when it turns out you get bad code out of that process. What can I say: we're flexible and we're deluded enough to think we can do everything.
|
# ? Nov 9, 2019 05:03 |
|
Absurd Alhazred posted:Testing is another vital element that managers skimp on in many ways. I find it simultaneously hilarious and depressing that every single time I've said "yeah it's fine now and you can skimp on that but if you don't fix it you're going to spend more down the road. Maybe next year, maybe not, but that's going to be an expensive mess to clean up" it's come true.
|
# ? Nov 9, 2019 07:50 |
|
Absurd Alhazred posted:Testing is another vital element that managers skimp on in many ways. I once made the mistake of breaking down my estimate for time to fix something (with a non-technical manager) in a way that included testing time and got the incredulous response "why does it need testing?"
|
# ? Nov 9, 2019 09:19 |
|
These new Lego sets are getting out of hand.
|
# ? Nov 9, 2019 10:02 |
|
Somfin posted:I once made the mistake of breaking down my estimate for time to fix something (with a non-technical manager) in a way that included testing time and got the incredulous response "why does it need testing?" If you were a good programmer you'd just write good code that doesn't need testing. Going to fire you and hire the first programmer i can find who can assure me they never test their code, not ever.
|
# ? Nov 9, 2019 10:51 |
|
Testing is not a value‐adding process.
|
# ? Nov 9, 2019 11:32 |
|
Negrostrike posted:Housii Housapodes
|
# ? Nov 9, 2019 11:35 |
|
Platystemon posted:Testing is not a value‐adding process. Triggered It even happens in manufacturing. I work in an industry that is intolerant of failure. This means that any changes are incremental and incredibly slow to develop. In a bid to speed development things up management decided to take a page out of software development, with the idea that you make some big assumptions early and test them - the implication being you might fail, but you'll learn a lot earlier. "Take risks" is even in the process map. We were wary, but went along with it because hey it'll allow big jumps. Well, we did that, and the part exploded, but we knew what happened and how to fix it... but management just saw a failed part and assumed we had no idea what we were doing. Even worse, one guy told our ultimate customer during a visit that "this entire test was ill-conceived" and did everything he could to throw us under the bus. This same guy did the equivalent of pointing at our flagship product, the equivalent of a high-performance top fuel dragster, and ask (in front of the customer) if it was [like a Model T]
|
# ? Nov 9, 2019 16:04 |
|
Covering your rear end with massive paper trails is extremely important. Getting the bosses to acknowledge your misgivings in writing and sign off on the idiocy anyway can literally save your career.
|
# ? Nov 9, 2019 16:51 |
|
Somfin posted:I once made the mistake of breaking down my estimate for time to fix something (with a non-technical manager) in a way that included testing time and got the incredulous response "why does it need testing?" That is a mistake. Testing is not separable from writing the code. That’s like trying to break out “change control flow” or “think about algorithm”. Testing is pervasive throughout development, even if it’s informally structured. Creating artifacts to assist with future testing can sometimes be distinctly analyzed, but only if you assume that “figure out what’s important to test” isn’t a meaningful endeavour.
|
# ? Nov 9, 2019 17:19 |
|
Subjunctive posted:That is a mistake. Testing is not separable from writing the code. That’s like trying to break out “change control flow” or “think about algorithm”. Testing is pervasive throughout development, even if it’s informally structured. Creating artifacts to assist with future testing can sometimes be distinctly analyzed, but only if you assume that “figure out what’s important to test” isn’t a meaningful endeavour. Seems somebody hasn't heard the good news of the waterfall model.
|
# ? Nov 9, 2019 17:33 |
|
klafbang posted:Seems somebody hasn't heard the good news of the waterfall model. the waterfall model does not contain good news!
|
# ? Nov 9, 2019 17:35 |
|
|
# ? Nov 10, 2019 03:50 |
|
Those sure are some error bars.
|
# ? Nov 10, 2019 03:57 |
|
How do internet people keep taking good ideas like "maybe I should look at my phone less" and building them into something that makes me hate humanity
|
# ? Nov 10, 2019 04:02 |
|
There certainly are people who derive pleasure from CBT, that's for sure.
|
# ? Nov 10, 2019 04:25 |
|
Paladinus posted:There certainly are people who derive pleasure from CBT, that's for sure.
|
# ? Nov 10, 2019 07:15 |
|
calcified beetle thorax
|
# ? Nov 10, 2019 17:41 |
|
|
# ? Apr 28, 2024 13:24 |
|
Paladinus posted:There certainly are people who derive pleasure from CBT, that's for sure. Yeah but most of us only need one dose to make us stop doing... well, anything, really.
|
# ? Nov 11, 2019 04:26 |