« first day (1137 days earlier)      last day (3856 days later) » 
03:00 - 20:0020:00 - 00:00

user20683
3:15 AM
@Clash Looks good to me and my English major girlfriend, do you want to farm that out to the Programmers blog as well?
 
9:39 AM
@WorldEngineer thank you and your girlfriend! What do you mean farm out to the Programmers blog? Does that mean I could post it to the programmers blog? That'd be awesome! Tell me what I have to do!
 
 
5 hours later…
2:49 PM
@Clash Cool article, prolog is definitely a neat concept to know, one of those things that could nicely solve some problems that would be an enormous pain in the ass in imperative code. Good tool to be spreading some knowledge on.
Maybe they just like gorillas, and they think that the ability to stay warm in the winter is preferable to sharks ability to regrow teeth. Did you ever think about that? — Jimmy Hoffa 5 secs ago
 
user55340
@JimmyHoffa That took multiple passes to parse... though maybe its just Monday.
 
@MichaelT Early on a monday as well
 
user55340
Quiet day in the office here. For some reason, local schools close on Columbus day - so the people who have kids where the other spouse has a less flexible schedule and works during the day are at home.
 
3:05 PM
@MichaelT Is today columbus day?
 
user55340
Yep. Didn't you know? Federal Government is closed today...
 
user55340
Its also Canadian Thanksgiving.
 
@MichaelT ...nyuck nyuck nyuck
 
user55340
(from last year weregeek.com/2012/10/08 )
 
user55340
 
user55340
3:09 PM
The idea of turkey + bacon + maple syrup seems interesting.
 
But if it is in fact columbus day today, let me express my disdain for the fact that we celebrate that terrible fuck instead of the scholar Amerigo Vespucci (though who knows what atrocities he committed in his explorations, probably not too many since his focus was on navigation and cartography more than anything else)
To be fair we really shouldn't celebrate anyone from that time period; they were pretty much all a bunch of shites.
In fact let's see here..
Here maybe we should celebrate this dude instead, he was born today, he's a signatory on the declaration of independence and has a great name to boot: Francis Lightfoot Lee
Or william penn, he may not have found the states, but he founded Pennsylvania... ok on second thought maybe not, we definitely have better states than that one..
 
user55340
3:24 PM
I look at it not so much a celebration of the individual, but rather the cultural heritage (italian americans - much like St. Patrick's day is about the Irish not St. Patrick) and raise awareness of the time and the issues that still remain (native americans - no, not saying that they exist is an issue, but rather take the opportunity to celebrate their culture).
 
user55340
South Dakota doesn't celebrate Columbus day but instead Native American Day.
 
user55340
Apparently on a state/county level, Indigenous People's Day is celebrated by Berkeley, CA; Sebastopol, CA; Santa Cruz, CA; and Dane County (HA!), WI.
 
user55340
Apparently Denver too...
 
user55340
Indigenous People's Day (also known as Native American Day) is a holiday celebrated in various localities in the United States, begun as a counter-celebration to Columbus Day. The purpose of the day is to promote Native American culture and commemorate the history of Native American peoples. The celebration began in Berkeley, California and Denver, Colorado as an alternative to Columbus Day, which is listed as a federal holiday in the United States but is not observed as a state holiday in every state. Indigenous People's Day is usually held on the second Monday of October, coinciding wit...
 
@MichaelT That's not surprising, I didn't realize until meeting my wife how little most Americans are taught about native Americans, all of Denver and surrounding areas are ex-natives land so we learn a lot about it just as a part of learning the history of our area. In elementary school you go out into the prarie and build a teepee and all of that nonsense but all as a part of learning about the natives of the area, their history etc
 
user55340
3:33 PM
(I wonder how the government shutdown affects this - one of the traditional-ish (stemming from the 1969 - 1971 takeover of Alcatraz) celebrations is to hold a sunrise ceremony on the island... as its NPS, is it open? Apparently they did get there this year too - iitc.org/…
 
@MichaelT Wisconsin had plenty of natives too actually, did you not learn a good deal about all of that in public school out there?
My wife growing up in Chicago well, they apparently spent a lot of time learning about the architecture and industry of the city instead heh
 
user55340
@JimmyHoffa A reasonable bit, though I also worked at a Boy Scout summer camp and taught the Indian lore merit badge... (and if you're familiar with OA, I was active in that too...). Though the 80's still weren't much for awareness... might be moreso now.
 
@MichaelT True, largely related to the times as well
 
user55340
The Alcatraz bit was a program I watched on KQED out in the bay area.
 
It was probably unheard of before the 90s for anybody to be taught Custard was a scumbag or that indians were so much "massacred" as opposed to "at war"
 
user55340
3:40 PM
I also try to have a more than average awareness of Native American issues in technology...
 
user55340
Things like why the Cherokee language is available on the iPhone as a built in language (despite (and because!) only having 10k native speakers).
 
I recall hearing on NPR about that and a couple other super rare old languages they're reviving through technology
 
user55340
(children learning it in school weren't using it on early sms phones - the nation petitioned Apple to add it to the iPhone (they had already added it to Mac OS since 2003) so that children can use cherokee outside of school)
 
is TesterMatingCycleLogger too long of a class name? Am I being too descriptive?
 
user55340
3:47 PM
@Sparticus For a moment there, I thought you were looking for the next name after Sparticus and wondering who's mating cycle you were testing and logging.
 
@Sparticus There's a faction of folks who argue variables and classes become too descriptive, but time has somewhat bore them out: Not many agree this is the case anymore short of ridiculous names. I would say that's not bad, though as a rule be creative with your naming to achieve as much succinctness and specificity as possible
 
Yeah, I'm logging the number of mating cycles for this particular testing board
 
user55340
(whee, confusing people on MSO by casting a CV on your own question with <3k rep...)
 
MatingCycleLogger seemed a little ambiguous
 
@Sparticus I'd use a word other than mating, not for the spurious reason but just because it's an uncommon word with non-specific meaning in technology
 
3:48 PM
It's pretty standard for my company
 
mainly because our minds are all in the gutter
I would agree it's pretty non standard
 
Then I would say the class name is fine so long as "Mating" has specific known meaning to you and yours.
presuming "Tester" is also a known thing too
 
Yep
 
Then you're fine. It's not MaterTestingCycleLoggingHybridConnectorForMacsWithUsbNoXeniaSupport
 
3:51 PM
Wait, that's bad? shit....
 
user55340
@JimmyHoffa You need to do it like the consultants I dealt with... MtrTstngCylLggngHybrdCnnctrFrMcsWthSbNXnSpprt
 
I'd like to buy a vowel
 
@Sparticus Clearly they programmed to The Wheel of Fortune rules: Vowels cost extra.
 
user55340
@Sparticus That will cost you 3 hours of time...
 
user55340
Hmm... Hangman with a dictionary of class and variable names.
 
user55340
3:53 PM
As an IDE plugin.
 
@MichaelT Too easy, you just put "Manager" in and "Singleton" and you've solved half of every class name
 
user55340
4:04 PM
Le sigh...
 
user55340
-5
Q: Write a program that will display on the screen the following using for loops: 345678345678345678

user104917Write a program that will display on the screen the following using for loops: 345678345678345678

 
user55340
Next time the Data Explorer is updated, I'm going to have to run a query on how many comments on P.SE and SO in the month of September and October were done that include the link to the open letter.
 
 
user55340
For the philosophical wine drinker...
 
user55340
 
user55340
4:16 PM
(this weekend I stumbled across a set of Oregon craft hard ciders...)
 
@gnat The weather outside right now supports the use of a singleton...
 
user55340
Out of close votes and its before noon... its going to be a long afternon...
 
...I want to buy a hard drive...
Anyone here any experience with MythTV?
 
4:33 PM
@JimmyHoffa don't buy 1 hard drive... buy 5.
RAID is pretty awesome
Especially RAID 5. Hooray for minimal parity
 
user55340
Raid 5 is a bit excessive for home use...
 
user55340
I've done raid 1 often (mirroring). Yea, its twice as expensive, but easier to set up and easier to recover from.
 
Yeah, and besides home purposes call me crazy but redundancy is totally irellevant, so I would raid 0 if anything
 
As opposed to? It's got the best storage/cost ratio
Yeah but raid 0 means that if one drive fails you lose ALL your data, not just the stuff on that one drive
 
user55340
@Sparticus Theres the 'storage/cost ratio' (lets call it SCR) and the "Administration : SRC" ratio.
 
4:35 PM
Most motherboards support raid 5 out of the box nowadays
 
user55340
@Sparticus But you need 3x drives for that.
 
I mean if you're on an IBM from the turn of the century you may have issues but my mobo supports up to 4 or 5 (I forget which) drives
3+ drives, yeah
but you get n-1 storage space
 
I habit to ensure anything of any value on my home machine is backed up online or in a variety of different places. There is minimal of value that ever goes on my home machine since it's only purpose is entertainment. And it's just a long-standing habit from many years of tinkering wherein I could at any moment jack something up and lose everything. I always use my home machine with the explicit thought that I may wish to completely reinstall everything on a moments notice
 
user55340
This is why Steam is great.
 
so to get 4tb of space, you can do raid 10 with four 2tb drives, or raid 5 with three 2tb drives
 
4:37 PM
@MichaelT !
 
(or you could do raid 1 with two 4tb drives, but raid 5 will still be even cheaper)
 
user55340
I'm looking at redoing my Mac's setup. Right now its 2x1TB + 1x 2TB + 1x 3TB... (more space as I upgrade). I'm looking at going through and redoing it as 4x 3TB (raid 1 on 2x of the drives - its not the best idea to have the system disk on a raid - even more of a pain to recover from).
 
I have a lot of media that would be a major pain in the arse to recover
No SSD for your system?
 
user55340
I'm more interested in storage space than speed.
 
I do it because then I can re-install at a moments notice (almost literally, it takes about 8 minutes to install windows 8 on an SSD) and maintain my data
4x 3tb on raid 5 - 9tb of usable, redundant data storage....
 
4:42 PM
@MichaelT I live dangerously (dangerously old) I've got 1x320gb HD right now
and a 1tb network attached drive which is pretty pleasant just sits on my router
 
Living life on the edge. Next you're gonna tell us its a raptor that spins at 15k rpm
 
my storage and media all goes on the network drive, so 320 is plenty for now, but as I'm setting up MythBuntu I'm pondering expanding my storage a bit
HD TV shows live recorded take up a lot of space
 
that's what got me into it. I don't have time to watch TV during regularly scheduled times
so I... acquire it and store it locally so I can watch it later.
 
@Sparticus Yeah, I haven't watched a tv show during a scheduled time in probably 6+ years
 
I also lack a DVR
and my roommates and I go through many TV series doing homework so I'm constantly bringing in new seasons
 
4:44 PM
Windows media center is pretty great for all that
 
Except if all of your media is in a format it doesn't recognize
then it's a royal PITA
It's either convert it to something windows knows, or use something else
my RPi is pretty awesome at that kind of stuff like that though. I have it running the XBMC build for RPi and it plays MKVs without a problem
 
I used it for a long time, but one of my tuners isn't supported since Vista and I just upgrade to Win 8 this weekend (MSDN subscriber downloads FTW), and am installing MythBuntu to try and get both of my tuners working at the same time.
...I don't even know why I'm bothering, my wife and I haven't watched anything that's on normal TV in years, we get everything from netflix and hulu...
 
mating_cycle_counter = TesterMatingCycleLogger(IM04TesterInfo.DUT_PROD_NUMBER)
over_the_limit_under_arrest = mating_cycle_counter.isCountBelowMax()
if over_the_limit_under_arrest != True:
    print "Maximum mating cycles exceeded. Please replace wiring harness to avoid errors."
is over_the_limit_under_arrest an appropriate variable name? I thought it fit rather nicely lol
 
@Sparticus I thought mating cycles caused arrest in the event of under_the_limit errors...
 
user55340
if over_the_limit_under_arrest != True: -- can't you write if !over_the_limit_under_arrest: ?
 
4:48 PM
here, let me fix it
 
user55340
though I'm not pythonic... but doing comparisons against boolean constants seems... excessive.
 
mating_cycle_counter = TesterMatingCycleLogger(IM04TesterInfo.DUT_PROD_NUMBER)
over_the_limit_under_arrest = mating_cycle_counter.isCountAboveMax()
if over_the_limit_under_arrest:
    print "Maximum mating cycles exceeded. Please replace wiring harness to avoid errors."
 
@Sparticus I'm not a fan of unnecessary intermediate steps, they're more acceptable if there's lots of them but in your case you don't have many... I would just write...
 
there we go. flipped the check in the actual method so it returned true if the mating cycle max was exceeded
 
if TesterMatingCycleLogger(IM04TesterInfo.DUT_PROD_NUMBER).isCountAboveMax():
    print "Maximum mating cycles exceeded. Please replace wiring harness to avoid errors."
 
4:49 PM
but then where do I get the chance to put someone under arrest
I need to hold the reference to the TesterMatingCycleLogger
we use it later
 
Ok
well if the result
but you don't need that intermediate variable
 
I could just put the isCountAboveMax in the same line as the if statement
yeah
59 secs ago, by Sparticus
but then where do I get the chance to put someone under arrest
What can I do to get more feedback from teammates during code review?
I suppose that's a question isnt it
Hmmm....
 
user55340
Culture thing there...
 
would that be off topic for programmers?
Don't see it anywhere on P.SE
 
@Sparticus Actually I don't think so, explain the scenario and current behaviours, and ask for approaches and techniques to elicit more critical/engaged participation in the code reviews
It would probably be a pretty great question. It's a culture problem like @MichaelT says, and a common one among many workplaces
 
user55340
4:55 PM
Previous employer, one team had an 'always green light code reivews' culture.
 
user55340
Didn't matter how bad the code was, team lead / manager let it go to production if it worked.
 
Well I just don't often get feedback on any of my reviews. Even little stuff like what Jimmy pointed out would be awesome
 
user55340
After a few releases, since reviewers knew their suggestions weren't going to be followed through, they just had a review of 'ok.'
 
I try and provide feedback like that on my teammates reviews, but I'm not sure if they even want that
 
user55340
35
Q: What does a code review look like?

GendoIkariI'm writing a code review process document for our team; we've never had a formal process in place although we do do some code review. I've found lots of articles talking about how important code review is, but I have one question in particular that I haven't found the answer to on the web... is...

 
4:58 PM
@MichaelT Yeah I worked one place where there were a hand full of engineers who were just like a voting ring for eachother, the reviews were required by process so I was rarely invited to many reviews. There came a point though where a couple of the teach leads started requiring developers to explicitly request me on their reviews so they had someone in the room questioning their approach
 
user55340
I recall doing one review... I had 5 pages of issues with the review... the code was from outsourced company and I was forced to give an 'ok' on the review (politics / billing stuff). We 'couldn't' pay them until they got an ok review, they weren't going to do more unless they got paid, code was going to be released the next release cycle...
 
@Sparticus This is very common. Many don't, but it's worth your while to do it anyway. Take it as an opportunity: Practice communicating these things in such a way as to be well-received. Giving a quality critique that doesn't leave the other party feeling somewhat bothered is a valueable skill and worth practicing.
 
user55340
(that was the company that did the "remove all the vowels" thing on long class names)
 
Me! Me! sarcasm aside, I've actually grown very annoyed of code reviews where there's no criticism of my code; I know I didn't do things perfectly and so lacking criticism makes me feel like I'm wasting my time even asking for the review. So I agree that nothing but criticism is bad, but none is as well. — Jimmy Hoffa Oct 11 '12 at 14:47
25
A: How important is positive feedback in code reviews?

Jimmy HoffaWhen I do code reviews I tend to just have a running monologue, so as I'm making sense of what I'm reading there will be a lot of "Ok, I see what that does.. Good it connects to this and calls that, alright.. and that piece depends on both of those alright.". I think in this way it's not "oo la ...

 
user55340
One team where we did start doing code reviews (once we had the time in the process and the tool to support it), we specifically wanted the jr. people to be looking at it and asking questions of the code. Its a learning experience.
 
user55340
5:03 PM
 
Yeah I'll often take code reviews on new projects to see what the hell is going on
 
user55340
Some people were in favor of the atlassian product line though - atlassian.com/software/crucible/overview (amusingly, the guy who was most pro atlassian left the previous employer to go to a company where it turns out they're 100% atlassian tool chain)
 
Hey, if you don't like what you're doing, change it.
 
user55340
Jira, Confluence, Crucible...
 
Admittedly it's not always that simple
but hey, can't blame a guy for trying
 
user55340
5:06 PM
Its more a "wow, I just went to a company that decided to invest in a toolchain from start to finish and it works so nicely!"
 
@MichaelT This is my first job running Jira/Confluence, Confluence isn't bad at all, Jira is on the otherhand not appreciated... Any toolset that needs that much cajoling to fit for you where the base-settings are not simple and general tends to breed poor process-adherence I find..
 
That's usually the experience people have with .net though
everything is so nicely integrated
I mean its easy to see why companies love it
 
user55340
@gnat hmm... I'll write a query that looks for questions closed for "resource" request and tries to find the ones that should get deleted sooner than later, but needs help getting there (cliff or delete votes)
 
user55340
@JimmyHoffa They've got a small team that deals with care and feeding of tools - something every tool chain needs to some degree.
 
@MichaelT Yeah, when you have the team that does that it works out a lot nicer, though it can become pretty bad, it just depends on the quality of that team: Too damn many fields in every ticket of any sort = bad adherence, this has been shewn out in every single place I've ever seen it (which is many)
 
5:11 PM
Hi all
 
user55340
@JimmyHoffa The thing that he liked was that the ticketing system and the code review system (and using Stash for code hosting in house) all tied together nicely. Doing a commit against an issue triggered the code review in the other tool and changed the status of the issue once the review was done...
 
@MichaelT Yeah that's the way the Microsoft stack has been for years
 
user55340
The key to the adherence is making the automation possible with as little work as possible.
 
It is nice being able to check in against a ticket which automatically kicks off a build running unit tests, updating the ticket with success or failure depending on how the build succeeded etc to move the ticket forward
 
user55340
So that its painless to do right, and painful to work around. Painful on both paths isn't good.
 
5:13 PM
@MichaelT Yeah, but painful on both paths is the vast majority
 
user55340
And thats where you need that care and feeding of tools team.
 
@jozefg howdy
 
user55340
To make the gears of the process line up properly and apply the proper lubricant.
 
I'm angry at Python... I should switch to Ruby..
 
user55340
@jozefg Um... you'll be angry at ruby soon enough.
 
user55340
5:14 PM
Switch to perl.
 
@MichaelT One place I worked that "care and feeding of tools team" decided "Yeah we'll make this work great! Most customizability and best quality process software will be- dun dun dun -hand written from scratch! It will have fields for everything imaginable so we can track everything really well! Yay process software we rule!"
 
@MichaelT So I have a stroke and than suddenly Python doesn't look so bad?
 
user55340
@JimmyHoffa The key is the C&F team isn't the one making the policies and decisions but rather acting on them... and sometimes saying no.
 
They literally wrote their own entire project ticketing/tracking system, a bunch of hand written visual studio extensions and plugins to control and complain wih relationships to tickets and permissions systems they had built into the ticketing system they wrote from scratch
 
.... I like ruby....
 
5:16 PM
(To be fair, the engineers throughout this entire company were the dimmest engineers I had ever worked with, ever, anywhere, ever. It was bad.)
@Sparticus How do you get the hash code for an object?
 
@JimmyHoffa jozefg.bitbucket.org/posts/… Remember that math answer I wrote a while ago?
 
Pretty sure you can just do .Hash() on the object
I've not run across that use case though
 
user55340
@Sparticus Such a statement is clear signs of brain damage.
 
I still think of perl as what happens when you put smalltalk on a train and have perl park a car on the tracks.
And maybe Lisp ran in to try to move the car and got hit as well.
 
5:20 PM
Alright, time to do some metaprogramming with python
 
user55340
> I started out as a BASIC programmer. Some people would say that I'm permanently damaged. Some people are undoubtedly right... But I'm not going to apologize for that. All language designers have their occasional idiosyncracies. I'm just better at it than most. -- Larry Wall
 
Sorry ruby not perl
 
user55340
(note the 'Hubris' attribute)
 
user55340
> Is LISP a candidate for a scripting language? While you can certainly write things rapidly in it, I cannot in good conscience call LISP a scripting language. By policy, LISP has never really catered to mere mortals... And, of course, mere mortals have never really forgiven LISP for not catering to them.
 
@Sparticus you just need to spend more time in the .NET stack. Having wandered language by language I have come to one fairly firm conclusion: (outside of smalltalk maybe) .NET is the best OO stack available by a long shot, over Ruby, Python, etc etc.
 
5:23 PM
.NET stack costs money that this poor college student does not have
(granted I have access to the latest .Net stack through MSDN for free, but that's another story...)
 
@Sparticus Since when? There's tons of open source folks out there living in the Express products
 
My stakeholders wanted it in RoR
thus we are in the Ruby Stack
doing crazy half html half ruby files...
 
@Sparticus I know that, thus why I said you will change your tune on that Ruby thing when you get to spend more time in the .NET stack
 
with a little CSS sprinkled in for funsies
 
@MichaelT Perl seems nice occasionally for just hacking stuff up, but whenever I try it it's like having a car with 4 wheels, but only 2 of them are where you think they are and the steering wheel will turn on the headlights if jiggled just so
 
5:24 PM
You know what you get for not using express products? A ton of enterprise stuff that is totally irellevant and unnecessary to people developing for themselves.
 
So I'm trying to test 4 inputs without repeating myself 4 times in a test suite
 
user55340
> Now, I'm not the only language designer with irrationalities. You can think of some languages to go with some of these things.
"Simple languages produce simple solutions" - C.
"If I wanted it fast, I'd write it in C" - That's almost a direct quote from the original awk page.
"I thought of a way to do it so it must be right" - That's obviously PHP. (laughter and applause)
"Everything is an object" - Smalltalk and its children. (whispered:) Ruby. (laughter)
"Everything is a hypothesis" - Prolog. (laughter)
 
I do an operation to one input and then look at all the others to make sure there was no cross talk, which is fairly simple
So I'm trying to do it dynamically so that it's more DRY
 
@MichaelT I hate that statement about Haskell, it is soooo inaccurate it's silly. You spend vastly more time writing functions in other languages than Haskell, if you wanted a cliche for Haskell it would be "Everything can be classified" or "Everything has a type"
 
that's not nearly as fun to laugh at Jimmy
 
5:28 PM
@Sparticus Show two of the code snippets
 
user55340
@JimmyHoffa Reading through en.wikiquote.org/wiki/Larry_Wall you can get an idea of the mind of the person who designed perl, and thus perl itself.
 
@Sparticus Yeah, but it causes a lot less confusion for people who eventually try to learn Haskell; it's one thing for a laugh line but it's another thing with how much the "everything is a function" moniker is actually said seriously all over the internet by people who don't know better
 
here's what I'm trying to do in C++:
 
err...
Let me put it somewhere actually readable
what does everyone use for sharing code snippets?
 
5:30 PM
@Sparticus gist.github.com is popular for that
then there's always pastebin and various typed pastes like jspaste, hpaste, etc etc
 
I should really start using github..
 
Man, it's really hard to describe the problem without you guys knowing a hell of a lot more about the problem.
I'm gonna just keep googling and hope I can figure it out. If I don't get it by the time I head home I'll try and excract the idea out of it
 
@Sparticus You have 4 tests which have near identical code right? That's not that hard to DRY up usually
...especially when you know how to use higher order functions which Python supports
 
Yes, 4 tests with near identical code. The only thing that changes is which input we are testing
I'm working on learning the higher oder functions
On each input, I run 4 tests, Pullup Test, Feedback Test, 32V Analog Test and 5V analog test
so there are 16 total methods that need to be defined
but the 4 pull up tests are nearly identical, the only difference is which one is the main one being tested, and which ones we check for isolation
 
Why don't you just have 4 methods with the "input" as a parameter? (then 16 methods that call those with their varied inputs)
 
5:37 PM
In essence, rewrite the test and leave holes where the data varies, making not of what should go in each whole. Than take the block, plop it in a function and make each of the wholes a parameter. Then you can make a list of what each parameter could be, and take the cartesian product and apply your new function
 
Err.... it's a limitation of the framework... that I made.... the name of the method is used as the test name
def DUT_Initialization_Test_With_Serialization(self, device_manager):
    """Initializes the DUT and serializes it if no serial number exists."""
 
@Sparticus Did we learn a lesson about tight coupling?
 
will show up as 'Running Dut Initialization Test With Serialization' on the screen.
 
There's a simpler solution here, don't test :) What could possibly go wrong?
 
@jozefg They have snow plows in your part of the country right? Trust me, you want @Sparticus to write these tests.
 
5:38 PM
I think the main lesson I learned is that frameworks are hard. period.
 
@Sparticus Yeah, that's true. Designing an API and designing behaviours is something that you learn from consuming a lot of them for a long time first and foremost. Seeing what other API designers did right and what they did wrong and how it plays out over time in a real implementation.
 
Oh I took a LOT from other testing frameworks
I have test suites and test runners and all that jazz
and made small... improvements. Or what I thought were improvements
like having the test being run show up on the screen
 
@JimmyHoffa Fair point... I like it better when I'm not being mauled by a snowplow.
 
@Sparticus I know, I'm saying after you've been writing code utilizing various frameworks for various purposes over the course of several years you'll get better at API design, it's one of few things that can't easily be quantified by logic alone but needs visceral experience to build up that knowledge of what does and doesn't work
or as you said, "frameworks are hard"
 
I mean DRY wasn't really a requirement of this project... so I could very feasibly just copy pasta for all the tests
it just doesn't meet my personal standards....
 
5:42 PM
@MichaelT well, so far your "marginal answers with links" query does this magic for me...
in The Water Cooler, Oct 3 at 16:24, by gnat
@RhysW side effect / additional motivation for this was when Yannis taught me to look into questions that "invite" crappy answers. Turned out that very often, positive score crappy answer just reflects equally crappy question, worth closing and deletion. When you see "here's the link" crap answer sitting at respectable +5, scroll to the top and typically there's "gimme link" crap question sitting right there, waiting for your vote
 
@Sparticus Can't you just write an overload of this function? (Does Python support overloading? @jozefg?)
 
@JimmyHoffa No, there's no types to overload on
 
I think my biggest problem is that each function needs to have it's own descriptive name
 
And arity based overloading is impossible with optional/variadic args
 
so it should be
Input_1_Pullup_Test
Input_2_Pullup_Test
etc
 
5:43 PM
@jozefg o.
 
@JimmyHoffa Yeah, it's arity based
 
yeah, arity. One of those fancy PL wonk terms I learned only after reading lots of crap about Haskell..
 
You can do multimethods reasonably pleasantly, there's a package somewhere on pip
 
I think the exec command is going to be used somehow
 
@JimmyHoffa My favorite word is profunctor
 
5:45 PM
@Sparticus don't hurtyourself now.
@jozefg Monoids are easy!
 
@Sparticus Erm.. I think your doing it wrong..
 
In the future when I see people complaining about how those nazi mods at Stack Overflow always close all those interesting questions, I'm just going to link them to this post.
 
@jozefg never! MY CODE IS PERFECT EVERY TIME! I AM SPARTICUS!
wait. Can I use the /__name__ attribute on a function to change it's name?
 
@JimmyHoffa They're nice too, I think a co-monoid would be useful too, a -> (a, a)
@Sparticus ish? It'll change how it shoes up when introspected, but how often do you introspect your tests?
Or is that your framework?
 
by introspect you mean something like this?:
def runTests(self, deviceManager, test_results):
        for test in self.TESTS:
            print "Running %s..." % test.__name__.replace('_', ' '),
            try:
                test(self, deviceManager)
                print "Pass!"
            except AssertionError, e:
                print "Fail!"
                print str(e)
                test_results.addError(str(e))
that third line counts as introspection right?
 
5:49 PM
@jozefg You just made me think of the meme. Guessing you haven't heard it. Just one of those strange rare things that are only found in the subculture of the subculture of the subculture that counts itself among people using words like "profunctor"
 
I forget where I heard the quote but it's to the effect of: No matter how deep you delve into a subject, there will ALWAYS be further sub divisions.
and I know that there is a relevant XKCD
 
@JimmyHoffa Thank you
 
@JimmyHoffa I live by the principle that if more than 10 people are in your community you ought to specialize
 
@jozefg Does this ensure you never have enough peers to compete with that you can only ever compete with yourself or something?
 
5:55 PM
@jozefg I am not an ant!
 
Category -> Cheerios
I like that substitution. You should create some TH library that allows precisely that.
 
@JimmyHoffa I like my Cheerios like a like my categories, small and free of russel's paradox.
 
@jozefg Because you prefer the name Zermelo? All I'm saying is that the word Schönfinkelization is a vastly underused term. If we aren't going to shy away from Profunctor why do we shy away from such a spectacular term as that?
 
@JimmyHoffa German > Category Theory.
In terms of confusing words at least
 
user55340
@gnat I'm thinking more looking at questions by close reason - identifying the 'resource' ones and helping those along faster.
 
6:05 PM
@Sparticus seriously though, you should be able to DRY that up real easily. Just tweak the framework or rename the hosting method so that the framework reads out "Running slew of tests" or something like that where the method name says "This isn't just one test"
 
@MichaelT I see. Would be interesting to have a query focused on these, I'd certainly give it a try
 
@MichaelT Larry Wall worked for NASA?
 
user55340
@JimmyHoffa Yep.
 
user55340
@JimmyHoffa Perl was originally a very small report scripting language for where awk + sed wasn't enough and you didn't want to delve into writing something from scratch in C.
 
@MichaelT Why in hells would NASA hire that nut job??
 
user55340
6:13 PM
This was at the heart of perl back then - perldoc.perl.org/functions/format.html
 
user55340
@JimmyHoffa Because he's a very intelligent nut job.
 
@MichaelT Yes, I get it, he knows regexp really well, but what the hell was that supposed to do for NASA?
 
user55340
@JimmyHoffa His regex comes from being a linguist and back in the days when CS was CS.
 
user55340
I think he was a sysadmin...
 
@JimmyHoffa Was he actually hired by NASA or by Caltech to work at JPL?
 
6:20 PM
@ThomasOwens Iduno, I was just reading some of those usenet quips of his from the link above and noticed the NASA email and then looked on wikipedia and see he "joined the NASA Jet Propulsion Laboratory after he finished graduate school."
 
user55340
@ThomasOwens He posted from "jpl-devvax.JPL.NASA.GOV" back in the day...
 
@MichaelT Contractors can get .gov and .mil email addresses. I mean, if that one crazy guy can get a job at MIT, I'm sure Larry Wall can be hired by Caltech to work at JPL.
 
user55340
I'm fairly sure he was a JPL employee... though I'd have to hunt down some more biographies for that.
 
@ThomasOwens "that one crazy guy can get a job at MIT" Two things: Which? and that "one" ?? I'm fairly certain that is a gross underestimate..
 
@MichaelT Wikipedia says he was. I don't know the relationship between JPL and NASA, but there are similar organizations that do work for someone but are run by others.
@JimmyHoffa Richard Stallman. I forgot his name.
 
6:26 PM
RMS was a professor at MIT?
 
user41796
Richard Matthew Stallman (born March 16, 1953), often known by his initials, RMS, is an American software freedom activist and computer programmer. He campaigns for software to be distributed in a manner, such that a user receiving it, likewise receives with it the freedoms to use, study, distribute and modify that software: software that ensures these freedoms (on receipt) is termed free software. He is best known for launching the GNU Project, founding the Free Software Foundation, developing the GNU Compiler Collection and GNU Emacs, and writing the GNU General Public License. Stallm...
 
user41796
I didn't know he was a pisces too
 
@JimmyHoffa Not a professor, I don't think.
He was employed by MIT, though. He has (had?) an office in the Stata Center.
The same place where Tim Berners-Lee and Noam Chomsky have offices.
 
user41796
> He considered staying on at Harvard, but instead Stallman decided to enroll as a graduate student at MIT. He ended his pursuit of a doctorate in physics after about one year (by[vague] October 1975[5]), to focus on his programming at the MIT AI Laboratory.[6]

While working (starting in 1975) as a research assistant[7] at MIT under Gerry Sussman, ...
 
I guess it's not that surprising, dude was unquestionably a treasure chest of brains, only problem was instead of having dubloons and jewels the chest was full of golden carapices and ruby encrusted jelly beans and equally unnecessary things
@GlenH7 Ok, that's more what I had recalled to be accurate, he worked in the AI lab, but that's not really "working for MIT", I think probably every grad student there "works in the AI lab"
 
6:30 PM
I actually found Berners-Lee and Chomsky's offices. But they weren't around.
 
user41796
@JimmyHoffa it's a notch above "student" but still several notches below even "adjunct professor"
 
user55340
> Someone told me that Larry a) moonlights as a moustache model, and b) has his own line of Hawaiian shirts.
 
user55340
 
6:46 PM
lolwut?
 
There are only two people in this world who can rock Hawaiian shirts: me and Bruce Campbell.
Oh. And the guy who works down the hall from me. So three.
 
user55340
@ThomasOwens Your profile pic is not hawaiian shirt... we need pics. Just like Yannis's hat.
 
It's getting too cold to wear one. I'll see if I have a pic when I get home. If not, I'll have to find a warm day to take one.
Looking now.
 
user41796
Thanks, and by fun, I meant "please delete posts"
 
user41796
@JimmyHoffa - yep, the race is on
 
6:53 PM
One doesn't appear to be spam. It answers the question and provides citations. Although it looks suspicious that he made an account only to post his links.
 
@ThomasOwens Also the principle manager I used to work for at a previous job: Some time in the mid 90s he decided he would never work another job that didn't allow full-time hawaiian shirts (he still wore them every single day circa 2009). He was also a cornell educated LISPer with many years of flight-simulator code under his belt so, he could probably pull of anything he wanted at any job he wanted.
 
user41796
@ThomasOwens It was the combination of the two answers targetting old posts without adding much / anything new
 
@ThomasOwens also note his PM.SE posts identically..
every post linking to a webpage that sells a book he wrote...
 
Yeah. I handled appropriately. I don't understand the excessive downvotes on the answer left. Although self-promotion isn't good, it does answer the qurstion.
 
user55340
@ThomasOwens spam flags carry with them an automatic -1 vote.
 
6:58 PM
Oh. Maybe it got flagged a lot of spam votes. But still...occasional linking to your own stuff isn't bad. Only when most of your posts do does it become a problem.
2/2 isn't a good start, though.
 
user55340
So, I'm not sure how many spam flags were cast on it.
 
user41796
@ThomasOwens I'm mulling that part over. I can see how it kind of does, but it's more about self-promotion than attempting to answer the book. And the "answer" is about as substantial as the saying on a chinese fortune cooking.
 
@GlenH7 "Man who stand on toilet high on pot; Read my book on the topic: www.timecube.com" ?
 
user41796
But I'll grant it's a poor question he's answering so it's going to be hard to generate a quality answer on a poor Q
 
user41796
I'll grant that his answers on Project Management are more substantial. So he answers the Qs there and gets his shameless plug in.
 
7:03 PM
just got two sudden successive bumps from the collider Q of last week, wonder if it was picked up somewhere..
@ThomasOwens Ash, housewares.
 
@ThomasOwens just for the record, he did the same at PM.SE, although audience there turned out warmer than ours. Per my recollection, he's quite a reputable author (though this didn't stop me at flagging the first post that was deleted - spam is spam, no matter what)
 
user55340
@GlenH7 Someone else commented with a MSO link, and I'm hoping to get +2 rep from the edit to remove it (didn't get any from the CW modification).
 
user55340
@gnat Advertise all one wants in the profile, make good material for people to want to check out the profile. But overt "shameless plug" in the post no matter how on topic... it detracts from the quality of the site.
 
0
Q: Is Stack Overflow a danger or benefit to collegiate CS classes?

KadinskiCollegiate courses in computer science are particularly tricky to balance due to how available answers to specific problems are online on sites such as Stack Overflow. For example, a student may look up how to properly sort a list in the language of their choice from a preexisting SO question, ra...

recommended reading: Open letter to students with homework problems -- "It is September once again (today is the 7316th day of September), and once again students are asking their homework problems on Stack Overflow and Programmers.SE..." — gnat 4 mins ago
 
user41796
@MichaelT There are two comments on the remaining answer now highlighting that issue for him.
 
7:10 PM
@MichaelT yeah. I think @GlenH7 nailed it...
Welcome to Programmers. While self-promotion can sometimes work on the site, it's better to directly answer the question first. Your answer dances around the main points of the question ("what is the main role" and "how is that different from...") as a run-up to your self-promotion. That leaves a poor impression of your answer, and can lead the community to believe this is just an attempt at spam. — GlenH7 3 mins ago
 
user41796
I tried to be nice.... :-)
 
@GlenH7 as far as I can tell, this time it worked out really well. :) Exposing the fundamental issues without sounding rude is sort of an art
 
user41796
An art at which I usually fail... :-)
 
@GlenH7 Have you tried an ad hominem approach? I find it successful towards many purposes.
 
user41796
@JimmyHoffa - nearly had to go down that route this AM with a review of a code review. <sigh>
 
7:26 PM
@GlenH7 If nothing else you can always go home after work, open a beer, and let open a verbal assault at your wall or the various inanimate objects you own. It's not quite as useful but it helps in it's own way too.
Plus there's few things that'll make the wife laugh like walking into a kitchen to see a grown ass man shouting that duck typing is no excuse for violating liskov substitution at a trash can
 
user41796
@JimmyHoffa Did that on Friday. Today was the calm version of that. Including where I opened up the IM history and showed "This is where I asked for XYZ. Why didn't you provide what I had asked for?"
 
@GlenH7 Ah, passive aggression, a fine choice for a monday morning in an office.
 
user41796
said another way, I had asked for baz() on a new class but not foo() or bar() out of two example classes I had provided. I got foo() and bar() instead.
 
tldr - don't fool with the bar or baz things will happen - fizzbuzz will bar the fool. — psr Aug 3 at 0:35
 
user41796
@JimmyHoffa polite, but not passive. Made it clear I had asked for baz() and there was a good reason why we needed just that.
 
user41796
7:31 PM
shouldn't that be "don't foo with the bar or baz things ..."
 
@GlenH7 @psr has MUMPS, forgive his addled mind
 
user41796
@JimmyHoffa I've got 99 problems and MUMPS ain't one.
 
user55340
@GlenH7 Having 99 problems when working in octal certainly is one of them...
 
user55340
"I've got 100 problems I need to get done today..." wow, thats a lot "yea, but I'm working in binary so if I get 1/hour done, I should be good..."
 
user41796
7:47 PM
@MichaelT If my 8 hour application run holds up, I'll be doing good too. Then I'll throw a 14 hour run at it. Stupid data integrity issues....
 
user55340
You're doing it wrong... it starts out as a 100 hour application, and the next run is 1000 hours.
 
user41796
kid you not - the delivery time I promised on this is "Christmas." And no, I wouldn't say which year. Apparently, one of my fellow devs has been yanking this product owner's chain for quite some time by only promising Christmas as the delivery time frame.
 
03:00 - 20:0020:00 - 00:00

« first day (1137 days earlier)      last day (3856 days later) »