On September 24 2015 00:02 Numy wrote: Lol some things in the real world are exactly like that, as someone doing Engineering with friends that are working as Engineers you'd be amazed MoonBear. Amazed. It's often far too hard to calculate exactly what is going on and the differences between the approximation and reality are too small to matter.
Actually find what you just said incredibly offensive and rude. So much of what science is is based on making the correct assumptions to get to your approximation. Most of the models we use to deal with micro systems are used because they give the best approximation as accurately determining the 100% real answer is just too hard to do. Just because you work in a field where that isn't a case don't go throw snide remarks at other people.
I don't see what's remotely offensive or rude about Moonbear's statement.
On September 24 2015 00:57 Alaric wrote: How to add 0.75 day to your trivial task: have to do some Javascript with it. Still as tedious to debug, and now I've got to learn how to make asynchronous Ajax calls wait for another one before launching, so more making mistakes and learning and debugging yay~
It's useful but I hate javascript. Even more than css.
On September 24 2015 01:55 Numy wrote: Siths don't like being called Jedis. Should really know the distinction by now, no?
Sith and Jedi also don't like being pluralized incorrectly.
Ah my mistake. I shall strive to improve my condescension until it's so good it's inoffensive no matter how incorrect it may be. Thus making sure to live up to the high expectations of the Sith and Jedi.
Magic later?
edit: Ok this is super cool bench. It's essentially like a power rack but doesn't require setup time and is a bit simplier to use. + Show Spoiler +
On September 24 2015 01:55 Numy wrote: Siths don't like being called Jedis. Should really know the distinction by now, no?
Sith and Jedi also don't like being pluralized incorrectly.
Ah my mistake. I shall strive to improve my condescension until it's so good it's inoffensive no matter how incorrect it may be. Thus making sure to live up to the high expectations of the Sith and Jedi.
Magic later?
edit: Ok this is super cool bench. It's essentially like a power rack but doesn't require setup time and is a bit simplier to use. + Show Spoiler +
That's pretty ingenious. I like the solution of moving the bench, makes so much sense. You haven't lived until you've done the roll of shame at least once though.
On September 23 2015 10:31 Lord Tolkien wrote: Question: anyone here seen Sense8 and can give me their review?
Been considering picking up Netflix to watch it but uh, not sure if wurf.
Very well made, haven't seen anything quite like it before. The first episode left me a bit bored because it was confusing and a bit slow but that more than definitely changed towards the end of it and later episodes. I would rate it very highly and the sheer amount of great things on netflix makes that an easy purchase.
Will do a more intricate review if you want me to.
While I think MoonBear and Numy are both correct, from my experience working at a Research Center, we had to make a lot of assumptions. Sometimes it was literally impossible to account for all of the variables as there were too many of them or we just didn't have a proper way to measure them accurately.
I think it really depends on what you're doing and what you want from the results. In some cases you're really just going to need the absolute precision and you can't afford any assumptions. Other times you just need more or less a general idea and you can make some assumptions.
On September 24 2015 03:14 Frudgey wrote: While I think MoonBear and Numy are both correct, from my experience working at a Research Center, we had to make a lot of assumptions. Sometimes it was literally impossible to account for all of the variables as there were too many of them or we just didn't have a proper way to measure them accurately.
I think it really depends on what you're doing and what you want from the results. In some cases you're really just going to need the absolute precision and you can't afford any assumptions. Other times you just need more or less a general idea and you can make some assumptions.
Yes, we come from vastly different fields that operate quite differently. My issue was with "Some things in the real world aren't like homework in class where you can Jedi hand-wave away appoximations with a disclaimer. Sometimes you actually need the right answer." It's condescending and implies that making assumptions with justification is something that is done in class which isn't the case at all. It's not "Jedi hand-wave away approximations", it's how real work gets done. It's also not just for "exploring unknown depths", it's for real current 60 year old figured out stuff.
I was working on a Factory last year for 6 months doing some consulting type work(More grunt work than anything :<) where I had do design some reaction vessels for the guy. I was trying to be too perfect in the work which landed up getting stuck a few times. The lead engineer comes to me and basically tells me to not worry so much about the smaller details and just get the overall thing within a reasonable percentage. It's just really important to document exactly what you are doing and why so when someone comes along later they can see what's going on. Friend working at Shell who I helped a bit on some diffusion issue made me laugh so hard. He had this huge issue because the work he was doing was basically fixing someone else that had gotten the wrong literature due to an incorrect assumption but the higher ups kept just telling him to work with it without bothering to read the literature.
But whatever. I was gonna drop it, I've said my peace. I'm clearly in the wrong anyway it seems.
On September 24 2015 00:57 Alaric wrote: How to add 0.75 day to your trivial task: have to do some Javascript with it. Still as tedious to debug, and now I've got to learn how to make asynchronous Ajax calls wait for another one before launching, so more making mistakes and learning and debugging yay~
It's useful but I hate javascript. Even more than css.
Are you using jQuery or another framework?
Prototype, for CakePHP. Still annoying to debug and stuff.
Confirmed, I can still do VLOOKUP()s. Unfortunately, the need for the LOOKUP() function lookup_array to be in ascending order didn't save me from having to use 20 nested if()s for a different formula...
I know there's got to be a more elegant solution, but I didn't feel like getting hung up on that piece of it for what is, more or less, a proof of concept.
On September 24 2015 03:14 Frudgey wrote: While I think MoonBear and Numy are both correct, from my experience working at a Research Center, we had to make a lot of assumptions. Sometimes it was literally impossible to account for all of the variables as there were too many of them or we just didn't have a proper way to measure them accurately.
I think it really depends on what you're doing and what you want from the results. In some cases you're really just going to need the absolute precision and you can't afford any assumptions. Other times you just need more or less a general idea and you can make some assumptions.
Yes, we come from vastly different fields that operate quite differently. My issue was with "Some things in the real world aren't like homework in class where you can Jedi hand-wave away appoximations with a disclaimer. Sometimes you actually need the right answer." It's condescending and implies that making assumptions with justification is something that is done in class which isn't the case at all. It's not "Jedi hand-wave away approximations", it's how real work gets done. It's also not just for "exploring unknown depths", it's for real current 60 year old figured out stuff.
I was working on a Factory last year for 6 months doing some consulting type work(More grunt work than anything :<) where I had do design some reaction vessels for the guy. I was trying to be too perfect in the work which landed up getting stuck a few times. The lead engineer comes to me and basically tells me to not worry so much about the smaller details and just get the overall thing within a reasonable percentage. It's just really important to document exactly what you are doing and why so when someone comes along later they can see what's going on. Friend working at Shell who I helped a bit on some diffusion issue made me laugh so hard. He had this huge issue because the work he was doing was basically fixing someone else that had gotten the wrong literature due to an incorrect assumption but the higher ups kept just telling him to work with it without bothering to read the literature.
But whatever. I was gonna drop it, I've said my peace. I'm clearly in the wrong anyway it seems.
If you really want to continue this, then hey sure. But this isn't some appeal to the middle where it's all "Oh both of you can be correct".
My position is that your choice of methodology will depend on two things. The first is the accuracy and precision required, the second is cost effectiveness. This is a very reasonable way of doing things. Yours is that there's no nuance at all and you can just do whatever because assumptions.
My view is that there is a range of approaches and you choose the right one for the task at hand. If you have two options A & B, and A is clearly better than B and can be done at no real extra cost you do A. Your arguement is that there's only ever one approach.
Heck your own example just supports my position. It wasn't that Lead felt that accuracy/precision wasn't necessary, but he was optimising for cost-efficiency. Why hire a consultant to do stuff when they can do the grunt work they were hired to and then let someone else tidy up the loose ends more cost efficiently. This is basically how consulting works? Push work down onto the junior resource and have them do the bulk work cost-efficiently and leave the tidying up to someone more senior who can finish it to the level of detail needed quicker and cheaper. It's how business works.
The lesson you learnt from your experience was that you can just do things more simply if you document it. Fine, if it works for you. But the real lesson is optimise for the task at hand. What are you really working towards? You thought you were optimising for a perfect solution and tried to do that. The firm you were working for just wanted a base outline where most of the time-consuming work was done by someone else and they would finish off the rest.
On September 24 2015 03:47 jcarlsoniv wrote: Confirmed, I can still do VLOOKUP()s. Unfortunately, the need for the LOOKUP() function lookup_array to be in ascending order didn't save me from having to use 20 nested if()s for a different formula...
I know there's got to be a more elegant solution, but I didn't feel like getting hung up on that piece of it for what is, more or less, a proof of concept.
It's so close I can taste it.
I don't know precisely what you're trying to do. But a general principle is that rather than creating a massive set of nested IF functions, try and instead identify the key states that matter. Set those as their own variables (or column variable, idk how your model works).
A technique I like to do is not even touch Excel and just draw out what I'm trying to do on a piece of paper as a flowchart. That flowchart represents the bajillion IFs you're trying to do. Then figure out if there are ways to simplify that flowchart such as removing redundencies, or definining states of the world where X+Y+Z are all true as one variable or something. (It feels a bit like linear network optimisation problems you do by hand sort of?)
Also some aspects might be easier if you use INDEX, MATCH, OFFSET, and named arrays. It also removes a lot of restrictions of LOOKUPS (needing the variable to be on the left is a big one for example and also means you can be more flexible with your primary key).
EPT
