hi, * how do you go about defining degrees of simplicity, so you can then talk about it with other people, and try to reach a consensus on how much to push for...
... a fine question and i don't have a well formed answer. i think simplicity is a subjective thing, and that people therefore need to figure out how to talk...
... A non-quantitative method would be - which one is easier to explain? I have found that that is almost always the better choice. This, of course, assumes...
* another metric for me is state space: the more conditionals you have, the worse the code is for many reasons: testing, maintenance, being able to explain the...
... thank you for the thought; that makes sense. what if i played devil's advocate for a moment and asked: are there cases where you can explain something...
I like that. I think, too often, that "simplest" is taken to mean "easiest to write", or "requiring the least effort to write", but I prefer to think about how...
Well, I would probably avoid the conditionals with the proper use of OO. But if I couldn't figure out any other way to do it, that might be the simplest way. I...
... having done more maintenance than blue-sky development, i agree to some large degree :-) so/but that means simplicity is subjective. i think making...
... i spoke poorly; i mean conditional in the sense of there being a large state space that is not handled by one set of code, but needs to have many pieces of...
Alan, ... In my experience, this hits the nail on the head "assumes that the person I am talking to understands..." Simplicity in communication is dependent on...
I sincerely agree with you, Matt (especially in the notion that communication involves both ends of the signal), but the very next question we are confronted...
hi, ... perusing the web and quotes, i'm currently thinking that there are several dimensions or metrics which sum up to simplicity. (a) you can't remove...
I do this exercise in class, which sort of addresses both questions, or at the very least provides some interesting results... I ask the students to imagine...
Scott, ... Beat them until they understand it? :) j/k ... the ... Agreed... ... to ... not ... I don't know. But I'll know it when I see it! Seriously...
... any favourite top 10s from that? :-) for me it can be hard to answer because 1) it depends heavily on the purpose and context of the code, and 2) often i...
... $0.02 i think the only sane approach is to start with your group, with an eye on what the wider community says and does, to try to help validate/inform...
Raoul, ... For my current position, I spent a half day pairing with some of the developers as part of the interview process. My personal opinion is that it...
... thanks; ive heard of places (e.g. mozilla, google, etc.) where interviews are not just a quick one day thing; you have to run the gauntlet, even if you are...
Sure... In no particular order, these usually appear on the whiteboard: 1) Can you test it? DID you test it, and if not, why not? (here we usually mean...
We start where we can start... ultimately, software development need to mature (in my opinion) into a real community, with standards and peer review and all...
Thank my students. :) It is a really fun, illuminating exercise, and very encouraging to find that I get *pretty much* the same answers (apart from the testing...
i don't mean to flood with links, but this one also resonated with me. http://wiki.hsr.ch/SimpleCode/files/SimpleSimplestOrSimplistic.pdf sincerely....
Hi, Raoul, I use Kent Beck's definition of simple design, which assumes test-driven development. Simple design: * does everything needed for now * passes all...
... of OO. ... be the ... large ... have ... Design patterns are about handling this problem. They are ways of isolating separate logics to work together...
... explain? ... understand ... signal... ... code. ... While I agree with you, I would say that this is where the industry needs to mature - we must all have...