The Whiteboard

12:22 AM

@psr well, how do you define that before? For example, my "epic fail" taught me that I need about 20 (better 30) hours just to refresh stuff I brag about in CV - that can easily make it a good month before interview when I have a full time job :)

Jimmy Hoffa

2:49 AM

@psr Yeah, I worked through that enterprise transition and did a lot of training to bring the others along into the wide world of LINQ et al, I know how much work it is. and if you've been out of it for a while it doesn't show, it's a distinct minority that know lock statements are just sugar or that finalizers take more generations (at least 1 more depending on the size of the FReachable queue the finalization thread may not execute it after the first collection)

@gnat Yeah, this is the crappy thing about interviews. I've always preferred coding interviews over trivia interviews for this reason, and when interviewing the trivia stuff I drill into just to see what's there but value shown problem solving way more

Not sure why so many people think coding in interviews really is tertiary, in my book that's at least 50% of your grade, it says whether you can do the job you're being interviewed for, the rest of the interview should just be used to find out if you can also do the next stuff that the job will morph into over 2, 5, 7 years

fen

6:11 AM

I sometimes write technical blog post with several interesting links (3...5) with my explanation/comments. Is this still popular, or maybe such "content curation" is not that popular....

gnat

8:26 AM

looking for advice. Re-reading my question made me wonder if it would be more appropriate to tag it as bug instead of feature-request? It's spelled like implementation doesn't match the spec: when I was a tester I submitted stuff like that as bugs

27

Q: In hotness formula, discard answers when voting evidence indicates that these are not good data points

In current version of “hot” questions formula (Qanswers * Qscore) *, all answers are assumed to equally contribute to question "hotness score", including even those downvoted into oblivion. Suggest to discard answers when there is a strong evidence that these do not provide good data points for ...

Ampt

12:55 PM

@GlenH7 Never got a chance to edit your post. Ended up playing phone tag with US Cellular for 2.5 hours over a 35 dollar activation fee that they charged me twice

by the end of it all I wanted to do was fire up deadspace and shoot some space mutants

Dynamic

@Weston.h Yes?

Yannis Rizos

Third day I'm back @ work, and I've already made someone cry... Yay!

Tweeted by yrizos on August 21, 2013 at 12:24 PM

Yannis Rizos

Didn't have time for a better picture (the SE hat arrived today), but here's my full collection of nerd hats:

Ampt

IS that a pontiac?

Yannis Rizos

1:14 PM

Heh, I wish. A SEAT Ibiza, coincidentally exactly the same as the red one parked in front (but mine's black)

Ampt

sometimes I forget you are foreign. I was like dude that sounds super exotic

gnat

@fen that way is unlikely to make it popular. As far as I can tell, days when links with short comments were hot are gone long ago, a more reliable way for tech blog to gain popularity is to invest substantial effort into the posts...

nngroup.com/articles/write-articles-not-blogs

as Nielsen puts it, "Write Articles, Not Blog Postings"

however that doesn't mean you should avoid writing these, only that the reason for this is not to gain popularity. Short "sketches", helpful "notes for self" could be useful as quick references, signposts to particular thoughts that may shoot later, eg when used as "seeds" / "building blocks" for more fundamental posts. Making "hit articles" is not the only (I would even argue not the main) purpose of the personal tech blog; it could / should also serve as an outboard brain

see eg My Blog, My Outboard Brain

Yannis Rizos

1:37 PM

@Ampt I'm not foreign, you are ;P

2

user41796

@YannisRizos quit abusing your mod status. "It's commonly known" that Greeks are foreigners.

user41796

Hence "it's all Greek to me"

2

user41796

(welcome back, btw)

Yannis Rizos

@GlenH7 No idea what you're talking about, it's all Greek to me.

3

user41796

@Ampt I've had those types of days recently too

user41796

1:41 PM

@YannisRizos Your diamond hat is pretty sweet. Not sure it's a better replacement than the "halo" on your current SE pic though.

Yannis Rizos

@GlenH7 I think I have that in the car as well - you never know when you'll just run off work and go to the beach.

user41796

Heck yeah. Brings back fond memories of a morning when my girlfriend at the time swung by my office and suggested we head down to Mexico and hang out for the day. That was a great, spontaneous trip.

Ampt

lol, if mine did that I'd have her checked out. It would take a day just to get to mexico from where I'm at

fen

@gnat thanks! I will review my ideas then. Maybe it would be better to post such links/articles through reddit and then comment, create a discussion.

gnat

@YannisRizos "well you wonder why I'm always driving black..."

►

user41796

1:53 PM

@Ampt I was living in Tucson AZ at the time, so it was a quick hop to head South to Nogales

user41796

that was back when the gangs were at relative peace with each other and didn't fight out in the open

Ampt

Then that Heisenberg had to go and ruin everything

No? No breaking bad appreciation in this room?

user41796

2:13 PM

@Ampt apparently it's no appreciation at all, not just with respect to breaking bad

Ampt

I'm just gonna make my own room then. With meth and walter whiteys

Jimmy Hoffa

@YannisRizos Why would you wish for a pontiac, greek cars don't break down enough for you?

Ampt

its true. They are seriously unreliable, especially the electronics

I feel like they've got Michael J Fox over there soldering their stuff together

thorsten müller

Reddit down for anybody else? Seems I'm doomed to get some work done today :(

Jimmy Hoffa

@thorstenmüller Don't do it! Here, thecodelesscode.com/case/1

thorsten müller

2:21 PM

@JimmyHoffa thanks but I read this already completely

Ah lucky me, the last one is new

Jimmy Hoffa

@thorstenmüller Last ditch resort, what-if.xkcd.com

user41796

@JimmyHoffa far higher quality content than reddit, no doubt

Jimmy Hoffa

(coincidentally, with that link I may have ensured others don't get any work done today as well)

user41796

bonus points to you then! :-)

thorsten müller

Must have been the Bradley Manning sentence that brought it down.

user41796

2:24 PM

My distraction of the AM will be playing with DE.SE

Ampt

Am I the only one who thinks that the codelesscode may be cheating in forcing people to draw their own conclusions from the philosophy there. People are always quicker to defend ideas that they feel they produced themselves.

user41796

@Ampt but that's how koans are supposed to work. A koan doesn't necessarily have an answer; it's there to make you reflect

Ampt

but it's really left to the author to make sure that what is left is impartial. If they guide you to almost an answer that takes, say, another 5 minutes to reach at the end

you'll believe that to be the right answer because you came to the conclusion on your own, despite the input of the koan

gnat

as far as I can tell, recent edits made this Q worth reopening (I'd even upvote it but am at limit today; have to wait till tomorrow)

-1

Q: How to evaluate Secure Authorization for server to client

I have a customer that wants me to build a web portal (Asp.net 4.0) that will communicate with a desktop client, tablet and/or smart phone (e.g. iOS and/or Android). I was thinking of using oAuth 2.0. But I'm not familiar with any potential cons to using OAuth. And I don't know how to compare O...

design authentication oauth2

shopping-dropping edits also managed to invalidate current answer, but that doesn't worry me since it's a low quality anyway (typical zero-effort shopping recommendation - "OAuth 2.0 is probably the most popular way blah blah")

1

A: How to evaluate Secure Authorization for server to client

OAuth 2.0 is probably the most popular way at this moment. For example many cloud storage providers use it in their APIs for 3rd party developers. You can find it documented on their pages - SkyDrive, Google Drive, Box (for example) - as their user authentication procedure. I guess if big compani...

user41796

@Weston.h is perhaps in a better place to explain how koans work as he's studied them as a form more than I have. I've just subjected myself to them. They do come in varying degrees of complexity and "obviousness" of what some conclusions may be

user41796

2:36 PM

@gnat thanks! :-) I've already voted to reopen that one too.

Jimmy Hoffa

@gnat Sorry, not convinced. Looks too "Evaluate the options for me without any requirements" which only grows into "Here's all the requirements, now architect it for me" to me. It's on the fringe but I don't see how an answer to the question in it's current form could be of any quality.

user41796

@Ampt I don't know that koans are meant to lead to a "right" conclusion either. They are there to help you reflect on what the right conclusion should be. Many a student has thought they knew the answer to which the teacher merely nodded "agreement" knowing that the student was still working it through

thorsten müller

Or hit him with a stick over the head regardless of what he answered...

Hmm, that's strange. This spam that was just deleted came from a 'real' account

Thomas Owens

Define "real" account.

thorsten müller

Moment, not sure, maybe I clicked something wrong (Otherwise 'real' would mean is older than a few minutes and has some rep and questions or answers)

Thomas Owens

2:45 PM

No, it was just made and was a network spammer.

One post on Programmers (and a few other sites) with spam.

thorsten müller

Yes, looks like I just got the first normal questions user.

We have this one relatively often the last days

Thomas Owens

It's been a network-wide issue.

Someone making accounts and blasting spam on multiple sites.

Jimmy Hoffa

@thorstenmüller I forgot you started learning the haskell, here this should keep you appropriately distracted: haskell.org/haskellwiki/Typeclassopedia

Ampt

@JimmyHoffa were you a librarian in a previous life?

thorsten müller

@JimmyHoffa thanks for the link, will have a look. Slowly things seem to make sense. Though while I basically understand what functors, monoids and monads are and how they work, I mostly fail to see why exactly they are THAT important. Must be my lack of functional thinking (I understand it far better on the level of folds)

Jimmy Hoffa

2:51 PM

@thorstenmüller Once you start threading computations together in a monad it will start making sense; it's the collapsing effect that you add another step to the computation and the monad returns you a single computation that has the whole rather than all of the computations strung together separately. The result being that upon adding a step to the computation you get something just as composable as the thing you started with

It'll sink in when you start doing it

@Ampt Good rule to keep in mind c2.com/cgi/wiki?TellDontAsk

thorsten müller

Well, at least I can read a lot of code now and get some idea what's going on. That's always a god point for learning a language. Plus I can write at least simple code of my own for stuff like the Euler problems. Since it's only a hobby thing I have all the time I want anyway.

gnat

@JimmyHoffa thanks, I'll take a closer look (since I didn't cast CV on it yet, I wouldn't mind to re-close if needed:). It just striked me that my comment pointing to issues in original revision, further replicated in close reason is now totally at odds with question text. In cases like this, I typically prefer reopen... even if it is further followed by re-close :)

Jimmy Hoffa

@thorstenmüller Yeah, but the cool thing about the monad stuff is you start seeing approaches to problems in the other languages too, any dynamic language with first class functions like your typical ruby could pull off monadic solutions just as an affect of the polymorphism available. Can you add functions to first class functions in ruby like in javascript?

i.e. in javascript you can add a function to Function.prototype and then all functions will have that function available to them where "this" is the function

user55340

@JimmyHoffa You're just trying to slip c2 in there to get him to read the rest of it?

Jimmy Hoffa

@MichaelT Nah, TellDontAsk is important, I'm helping!

@Ampt another good bit to ponder on cse.chalmers.se/~rjmh/afp-arrows.pdf

Ampt

3:03 PM

@JimmyHoffa my problem is that I could break it up into multiple classes, but they wouldn't be reusable, so all I get are smaller, but more numerous files.

sure, I could have an input class that would manage an input's behaviors

but that one input wouldn't necessarily (or realistically) mimic an input on a different board

the other thing is that this is supposed to provide an 'API-like' interface for our hardware engineers to be able to manipulate the devices more easily

Jimmy Hoffa

@Ampt And this is good. Smaller is more manageable, more maintainable, more readable.

SRP

Breaking things apart isn't only for reuse, you break things apart to make them easier to mentally parse down the road, and breaking things apart makes your whole system easier to understand from a higher level view point.

Ampt

I'm not positive that I gain enough from SRP to warrant it's use in this situation. My methods are small. 2-3 lines. its basically filling out a form so that a single method call does something tangible

def setHardwareRevision(self, newHardRev):
s= SDO(self.SDOAddress, GPIO.HARDWARE_REV, 0x00, newHardRev)
hardwareRev = CANBus.send(s.toString())
return hardwareRev

def getSoftwareRevision(self):
s= SDO(self.SDOAddress, GPIO.HARDWARE_REV, 0x00)
hardwareRev = CANBus.send(s.toString())
return hardwareRev

Jimmy Hoffa

@Ampt A system's appropriate level of rationalization is entirely up to the author's taste

Ampt

those are some example methods

less indents apparently

thorsten müller

@JimmyHoffa interesting question about Ruby. While I'm quite comfortable with Ruby I don't use those meta programming parts that often. A everyday function is not an object as such, though you can convert it to a Proc object. This makes the answer a bit difficult but I would say it's more a No than a Yes.

Ampt

3:08 PM

its just a lot of those

Jimmy Hoffa

@Ampt Sounds like procedural code, you're tempted to call this module or class or whatever it is "Utilities" aren't you? That should be your first tip-off that something's a foot.

Ampt

yes, I would define it as a utility

is there something better I should be doing?

I looked into masking these method calls with variables so you would just do stuff like gpio.hardwareRev = 0x1234

and it would use the set method on the back end

but I didn't think that we gained anything as opposed to gpio.setHardwareRev(0x1234)

Jimmy Hoffa

@Ampt Also those are a lot of copy and paste, you should have one implementation of that i.e.

def setHardware(self, gpio, offset, val):
    return CANBus.send(SDO(self.SDOAddress, gpio, offset, val).toString())

def getHardware(self, gpio, offset):
    return CANBus.send(SDO(self.SDOAddress, gpio, offset).toString())

Ampt

ah, but these methods won't repeat. Maybe hardware rev

but not every device will support a hardware rev

Jimmy Hoffa

@Ampt I didn't hardcode hardware rev

the GPIO and offset are simply parameters

Ampt

3:14 PM

ok, but then the person coding it still has to know the underlying CAN implementation; that is the offset

user55340

Any SO 10k users (or mods) want to poke at...

user55340

1

Q: Move question to programmers.stackexchange.com

Shouldn't this question have been moved to programmers instead of being deleted? http://stackoverflow.com/questions/38144/how-can-i-convince-skeptical-management-and-colleagues-to-allow-refactoring-of-a

discussion migration

Jimmy Hoffa

@Ampt those should be consts

named consts not magic numbers

Ampt

as opposed to methods?

right now they are methods which use consts

Jimmy Hoffa

then all the user needs to do is pass in the appropriate enum or const or whatever

when I say const I mean named

Ampt

3:15 PM

named consts

the 0x00 is a placeholder for testing

that would be something like SUBINDEX_ZERO

Jimmy Hoffa

Well there ya go

Ampt

but GPIO.HARDWARE_REV is a const that represents the index of that can message

Jimmy Hoffa

you just simplified all those methods down to two, and it'll be easier to maintain as some devices won't have all the can messages

so the consumers can choose to call with the can messages that the devices do have, and add new ones to the GPIO list as they show up without you needing to add methods

Ampt

the messages stay very, very stable

Jimmy Hoffa

they just add the can message index const to GPIO and they can use the general set/get methods without you having to write new set/get methods

Doesn't mean you should create lots of unnecessary code just as get/set accessors when they all duplicate

If you want to use get/set accessors, you should at least write them to call the generic method rather than copypasta

Ampt

3:18 PM

ok, point taken there

still not convinced that moving away from having a long list of methods available for a device. The idea is that this is painfully simple to use for someone who doesn't know, and doesn't want to know anything about CAN

they want to turn a pin on the board on, they can do it with one method call

that, ideally, would have documentation in the form of doxygen

method name, inputs, results all spelled out

because we could have our technicians editing these scripts potentially

which would be really, really helpful

and they aren't stupid, but they don't have any need to know anything about CAN networks or anything like that. All they know is that an output or input isnt behaving as expected, so I can provide a better set of tools for them to be able to test it. Maybe switch it on and off really fast, or set it on and hold it until they say to stop

etc etc

user55340

conversation at work... someone misheard "EMP blast" as "MP3 blast", hilarity ensues.

Ampt

EVERYBODY HIT THE DECK! ITS A LOW BITRATE MEDIA FILE!!!

user41796

4:19 PM

Random Q - is there a term for using an abbreviation of an class / object instead of naming it for what it's supposed to do? Java has it all over the place, and it's commonly used when you need a "quickie" variable to do some work in a short method.

user41796

StringBuilder sb = new StringBuilder()
instead of
StringBuilder output = new StringBuilder()

user41796

I'm asking because I noticed one of our newer team members using that approach in a way that obfuscates what the code ought to be doing.

gnat

@GlenH7 Systems Hungarian notation?

Hungarian notation

Hungarian notation is an identifier naming convention in computer programming, in which the name of a variable or function indicates its type or intended use. There are two types of Hungarian notation: Systems Hungarian notation and Apps Hungarian notation. Hungarian notation was designed to be language-independent, and found its first major use with the BCPL programming language. Because BCPL has no data types other than the machine word, nothing in the language itself helps a programmer remember variables' types. Hungarian notation aims to remedy this by providing the programmer with e...

user41796

@gnat that's close - except in this case it's not including any information about the purpose of the variable.

gnat

Systms hungarian, cut to the prefix

> In Systems Hungarian notation, the prefix encodes the actual data type

user41796

4:24 PM

right now I'm thinking it's just lazy. :-)

user55340

@GlenH7 if its just a temp variable with a limited scope, I wouldn't mind (and occasionaly do) just use StringBuilder sb = new StringBuilder();

gnat

in Java, stuff like StringBuilder sb is rather idiomatic for short-term use, so I would be careful calling it obfuscation without first checking scope of it. I'd say usage spread over 32 lines apart is rather criminal, one spread over 3-4 lines is "legal"... legal in sence...

user55340

Now, you might want to talk to the coder about properly initializing the StringBuilder.

user41796

@MichaelT with the restrictions you applied, I agree. But he's not sticking to those limits

gnat

in sense weird but traditional and considered easy to read

user55340

4:27 PM

@GlenH7 if its not limited scope or spans more than an eyeball full, then smack him.

gnat

and actually changin the idiom in these cases makes code harder to read

user55340

And encourage him to use either new StringBuilder(init capacity) or new StringBuilder(String str)

user41796

@gnat agreed, and I think the length or complexity of the functions is part of the problem

user55340

sb = new StringBuilder(); sb.append("foo"); is wasteful. It should be sb = new StringBuilder("foo");

user55340

Or if you know you're going to be building a 1kb long string, sb = new StringBuilder(1024);

gnat

4:29 PM

@GlenH7 yup. You just have to find a compelling justification (preferably with reference to authoritative sources) that wide-span usage considered harmful)...

one minute I'll check Java code conv it might be there

user41796

@gnat and that's where I was going to go next if I knew the term to google-fu

user41796

@gnat Thanks, I knew it's done a lot in Java, but didn't know if it was official convention

gnat

oracle.com/technetwork/java/codeconventions-135099.html

Except for variables, all instance, class, and class constants are in mixed case with a lowercase first letter. Internal words start with capital letters. Variable names should not start with underscore _ or dollar sign $ characters, even though both are allowed.

Variable names should be short yet meaningful. The choice of a variable name should be mnemonic- that is, designed to indicate to the casual observer the intent of its use. One-character variable names should be avoided except for temporary "throwaway" variables. Common names for temporary variables are i, j, k, m, and n for integ…

user41796

He's pedantic (like all programmers) and will point out he's using two characters

gnat

that's sufficient to smack the guy, you just need to prepare to defend from braindead attacks like "oh it says one-char but I have two"

user41796

4:32 PM

here's another example: var cps in results

user41796

whereas I'd be good with var result in results

user41796

cps happens to refer to the class that results is a List<> of.

gnat

@GlenH7 see above, be prepared

...push more to "The choice of a variable name should be mnemonic- that is, designed to indicate to the casual observer the intent of its use."

user41796

@gnat I think that's a good approach

user41796

@MichaelT I'll point out the string builder initialization as well. That's not the least efficient portion of his code though.

thorsten müller

4:44 PM

How can there be a user without a link to his profile?

0

Q: Breaking down a complex story at project start

I'm trying to get to grips with agile project management (with Pivotal Tracker) but keep finding myself running into walls when trying to define the first few stories of a project. Take for example this very simple story: "A user should be able to tag a product" Assuming I've already defined "p...

scrum

user55340

@GlenH7 Its one of those things. I once dug into it and looked at building a 1kb string with 256x calls of 8 character appends and how it grew the backing array. It was painful.

gnat

@GlenH7 another possible braindead attack to be prepared to, is, oh but it's old "Revised April 20, 1999". For that, first point out it's published at modern site, and there's no mention of it being outdated. And, to smash him harder, add like so freaking what, JVM instruction set didn't change in between 1996 and 2011, they just keep stuff that works as long as it works

user41796

@MichaelT I recently got burned on it too. We generally don't init our SBs. However, I was cleaning up some code where the SBs had gotten so large that they were leaving on the large heap

user41796

@gnat fortunately I can pull rank to a degree there. I have more seniority on the team and he recently hired on

gnat

@thorstenmüller question migrated from SO and asker didn't (yet) registered here

user41796

4:48 PM

so I would phrase it as a "this is our undocumented convention ..."

gnat

@GlenH7 not quite good idea in my experience. Rather add this at some team wiki. Like naming conventions as per Java Code Conv. Newcomer users that tend to deviate from undocumented conventions make a great engine to finally document these :)

user41796

@gnat I'm kind of loathe to go down that route as well. My point was more that I could if I absolutely had to. We have a standards doc that we follow, but this isn't addressed in there. Perhaps I should present that task to him so he can become more familiar with our standards....

Ampt

Woohoo. Build is finally done after ... looks at watch 113 minutes!

Jimmy Hoffa

@MichaelT I used to do this, I usually catch myself anymore, even iterators in a foreach or indexes in a for loop I give meaningful names anymore like for(int productIndex = 0; productIndex < products.length; productIndex++)

user41796

@JimmyHoffa the problem is when you have to enhance the code to do something else. What was once a nice, ephemeral variable has now taken on meaning

Ampt

5:02 PM

I usually just use I. I guess I'm guilty of this as well

and sb would be the name of my string builder if I only used it for a few lines

although I would probably only do that in java. It seems to me that java enjoys making objects out of just about everything

and I'm only so original when it comes to giving a name to every line of code

user55340

i as a loop variable has a long history back to the days of Fortran.

user41796

@Ampt that's the difficulty I'm having - how to draw the line between "short term use okay" versus "lengthy function use not okay"

Jimmy Hoffa

@GlenH7 If he's a recent hire and you have to pull rank rather than him just appreciating your critique as what's expected of his new employer; that's pretty bad form on his part.

user55340

If you can cover it up with an index card on the screen, ok. If it takes an 8x10", bad.

user41796

@MichaelT these are 25 - 40 line methods. Right at the edge, IMO

Ampt

5:06 PM

@JimmyHoffa Ok, but it's not like he knew that it was wrong. Theres nothing telling him that it's wrong other than experience of an older member, which, while a great indicator, isn't always 100% truth

Jimmy Hoffa

@GlenH7 Perhaps you should present to the greater committee an addendum to the coding standards

Ampt

I would suggest making a change to the coding standard if you are going to enforce it here

user55340

156

A: Is it OK to split long functions and methods into smaller ones even though they won't be called by anything else?

Testing code that does lots of things is difficult. Debugging code that does lots of things is difficult. The solution to both of these problems is to write code that doesn't do lots of things. Write each function so that it does one thing and only one thing. This makes them easy to test with...

user55340

What is the cyclomatic complexity of the method?

Jimmy Hoffa

@GlenH7 Which is good! It'll make you think "Shit I'm indexing arguments with productIndex, this isn't right; I better make sure I'm validating my loop condition against arguments.length and not products.length now..."

user41796

5:08 PM

@MichaelT my version of VStudio doesn't calculate that apparently

Jimmy Hoffa

@GlenH7 There are tools

4

Q: Best tool to determine code Cyclomatic complexity

Can people suggest the best tool to determine the cyclic complexity with in a C# winforms code base.

c# cyclomatic-complexity

user41796

@Ampt Except in this case ... it is always 101% truth.

Ampt

@GlenH7 Obviously this is the exception to the rule master.

Jimmy Hoffa

@GlenH7 How cute, the college kid thinks truth is somehow related to how he should or shouldn't write his code in the industry, he's in for a rude awakening...

user55340

@GlenH7 Different approach - how many if and loops are there?

Jimmy Hoffa

5:10 PM

;P @Ampt if you're new somewhere, you follow their idioms before you try and enhance them; because when you're new to a place you don't know what you don't know

user41796

@MichaelT usually at least one loop, sometimes two. Minimum of 2 or 3 if / else statements

Jimmy Hoffa

@MichaelT how much nesting is a better indicator

@GlenH7 HE'S A WITCH! BURN HIM!

Ampt

@JimmyHoffa thecodelesscode.com/case/2 How can one know the idioms without being told master? I must meditate more to learn this method of mind reading.

@JimmyHoffa Woah woah woah, lets not jump to conclusions here. Where did we leave the duck?

Jimmy Hoffa

@Ampt Glen will tell him, I'm speaking to when you are told of a company's idioms that you just joined, you acquiesce, that doesn't mean you don't present thoughts and have a conversation, just that if they wish you to do it one way, that's the way you do it

user41796

(now I'm really ranting) he also is using var a lot. var has it's uses, don't get me wrong, but it shouldn't obscure what the underlying object is. You should know by looking at the code and without thinking too hard what should be coming back on that var

user55340

5:13 PM

@GlenH7 A rough ballpark for cyclo complexity is: count if, case, while, and for. Generally, 1-4 is low complexity, 5-7 indicates moderate complexity, 8-10 is high complexity, and 11+ is very high complexity.

user55340

I try to keep it to 7 or less.

Ampt

@JimmyHoffa OK, but no where (from what I've read here, I may be wrong) was it written that you should avoid 2 letter names for short term use variables.

Jimmy Hoffa

@GlenH7 I hated var until I learned Haskell. Inference is your friend

but overuse is still bad

user55340

@Ampt A discussion between Monty Python fans will always devolve into just quoting lines to each other.

user41796

@MichaelT these are hitting moderate then.

Jimmy Hoffa

5:15 PM

I use it in the case of var bla = new Bla() or var prooduct = myFunObj.GetFirstProoduct(); because in both cases it's clear what type you're getting

Ampt

He's using it as the class? BURN THE HERETIC

@MichaelT I'm not seeing the problem here.

user55340

As an aside, this is why automated source code analysis as part of a CI build environment is a good thing.

Ampt

@MichaelT psh, back in my day we made silly statements and we looked foolish

user55340

@Ampt And we used lint too.

user55340

Lint (software)

In computer programming, lint was the name originally given to a particular program that flagged some suspicious and non-portable constructs (likely to be bugs) in C language source code. The term is now applied generically to tools that flag suspicious usage in software written in any computer language. The term lint-like behavior is sometimes applied to the process of flagging suspicious language usage. Lint-like tools generally perform static analysis of source code. Lint as a term can also refer more broadly to syntactic discrepancies in general, especially in interpreted languages...

user55340

5:18 PM

> Stephen Johnson. Lint, a C program checker. Computer Science Technical Report 65, Bell Laboratories, December 1977.

Jimmy Hoffa

> The answer to life, the universe, and everything. Some report 42, Bell Laboratories, Sometime 1974

I'm pretty sure that actually happened from my understanding of bell labs

Ampt

One of my profs worked at bell labs during their hay-day. (he was actually also the dept. head) I'll honestly state that he was one of the worst teachers I've had

not to lessen what he did

Jimmy Hoffa

Bell Labs was the Google of its day

Ampt

oh, trust me, I know. He couldn't stop talking about it

@GlenH7 have we decided to burn him or not?

gnat

@JimmyHoffa I wouldn't bet it's a bad form, could be just juniority. For junior devs it's rather normal (and to certain extent beneficial) to challenge all regular ways; a tricky skill of authority appreciation when it makes sense comes with experience

Jimmy Hoffa

5:24 PM

@Ampt My vote is burn based on every method having on average 2-3 if/else blocks and one loop.

user41796

@Ampt he's a newbie (for our team) so he'll get the code review and I'll follow-up with him. Some of the structure he put in place was good. But he glossed over a few details that I need him to go back and resolve. Gave me the opportunity to point out the teams preferred approach

Jimmy Hoffa

@gnat Sure, from a junior it's expected. I'm no fan of blind acquiescence, but I guess the knowing when to listen and when to argue is as you said, something that comes with experience

@gnat but @GlenH7 is a real engineer, working for an engineering firm, they don't have juniors in places like that, I presumed he's at least a senior fake-engineer like me

Ampt

Confirmed. There are no junior level positions. Only senior positions and "senior" positions.

or interns

user41796

@JimmyHoffa Not true. The organization hires new grad every year. And we have hundreds of openings for both experienced and new engineers alike.

Jimmy Hoffa

@GlenH7 hundreds? How big is this org?

Consulting org?

user41796

5:28 PM

My particular team isn't that large (~10 of us), so we don't hire every year. One of our new hires was effectively a "new grad" and the other (today's subject) was an experienced hire

user41796

We're international. Local office(s) have ~3k employees. Worldwide is ~6k I think

user41796

And we're growing - targeting ~10k WW in 7 - 10 years.

Jimmy Hoffa

@GlenH7 Is it a good company? This has to be consulting, no?

user41796

@JimmyHoffa reasonably decent company yes. A bit stingy on pay and benefits, but I think that's a universal complaint.

Jimmy Hoffa

The only places that have that many software developers that aren't consulting firms are software companies we'd have heard of, your microsoft, google, etc

user41796

5:31 PM

As far as consulting, that's hard to answer. It's not IT consulting. But it is engineering consulting as well as management of projects.

Jimmy Hoffa

@GlenH7 Right, like thoughtworks

user41796

we perform a huge number of studies and guide our clients in making decisions based upon the engineering facts surrounding their question

Jimmy Hoffa

Or do you mean all-engineering as in physical, electrical, civical, histrionical (my personal favorite engineering discipline)

user55340

@GlenH7 Like how to test if a hard drive's heat makes it slower?

user41796

We're heavily involved in energy infrastructure, and we're involved in almost every facet of that from design to analysis to construction to enhancement

user41796

5:33 PM

@MichaelT At the power levels many of my employer's clients play at thermal resistivity is a very real effect and something that has to be accounted for

user41796

@JimmyHoffa Nope, but I'd rather not say who I work for. I have a degree of obligation to keep my professional activities separate from on-line activities. And / or I try to be very careful in how I'm presenting the "authority" of my experience. The joys of being a PE. Either way, it's definitely not a household name by any means.

Ampt

@GlenH7 you're talking about very long wires then right? Or are you really making enough change in resistance inside a computer to make a difference?

Jimmy Hoffa

@GlenH7 Yeah I figured as much, don't mean to prod. Either way that is interesting. (Do they have offices in Denver in case I ever cared?)

user55340

@GlenH7 eg its not ge

user41796

@Ampt transmission level power, yes. I'd say anything above 12, 13kV has to start worrying about thermal effects. High voltage transmission (100kV) it's a very real concern.

user41796

5:37 PM

@JimmyHoffa Indeed we do, although they focus on our Water utilities clients

Jimmy Hoffa

@GlenH7 Which makes perfect sense considering water is much more scarce than energy out here

Ampt

@GlenH7 oh ok, not data transfer, you're talking power transfer

user41796

Neat fact about 345kV transmission lines - substations that receive and transform those signals to other levels have to have the fencing surrounding the substation grounded. Otherwise the reactive energy from the 345kV line will light up the fence and you'll get a nasty shock.

user41796

@Ampt yes, although some people (not us) are playing with broadband over power lines.

Jimmy Hoffa

@GlenH7 Other neat fact about 345kV transmission lines - THEY WILL EFFING KILL YOU.

user41796

5:40 PM

@JimmyHoffa pretty much anything coming off of a substation will kill you. Subs run 12 - 13kV to the transformers on the poles, then step down from there to household levels. Breakers on substations are at the 400 - 600 Amp range. 12kV @ 400 Amp is gonna hurt bad.

Jimmy Hoffa

@GlenH7 Moral of the story, cyclomatic complexity in your case, could literally kill someone.

Ampt

@GlenH7 hurt? Would you even feel it at that point?

350 for 4 -2tb hard drives. That sound like a decent deal to anyone?

I need more space desperately and I'd love to have some redundancy

user41796

@Ampt Not sure on that. I used to work for the local power utility. We had a guy who got caught by a downed power line. He lived, but with some really nasty burns that left him disabled. I think the question is how long it would take to stop your heart and if it would affect nerve transmission or not

user41796

@Ampt If I had to guess, I would say it would hurt unbelievably bad for an exceptionally short period of time.

Jimmy Hoffa

@GlenH7 Your heart would definitely stop. The electricity would probably travel through your body quicker than the nerve impulses it was triggering, I suspect your brain would not interpret the message as pain because your brain would already have so many other activities going on just from responding to the jolt itself by the time those nerve impulses got to it

user20683

6:43 PM

@GlenH7 I've been shocked by an outlet before. It doesn't hurt for very long. Heart started racing though

Ampt

@Weston.h same. More scares you than anything else.

of course your average outlet isn't quite measured in kV...

Transcript for