The Whiteboard

user55340

12:00 AM

The thing is that the global state stuff that is so tempting and easy in an SE app becomes forbidden in an EE app... unless you really want it and know what you're doing.

user55340

(as we speak, I'm dabbling with a servlet .war that figures out where it is deployed to (is it /foo? or /bar?) and then goes and reads some configuration from couchdb and puts that into the application context for the rest of the servlets to be able to read)

user55340

You'll find annotations in lots of places with Java EE. Its how the libraries and containers know about the whats in the app itself.

user55340

And yea... get to know and love maven. You will after a bit. Well, maybe not love, but for EE stuff the structure it imposes and the dependancies its manages are better than the alternatives.

user55340

You might also glance at gradle... but thats yet another one.

user20683

"Real" Programmers

user55340

12:11 AM

@WorldEngineer A classic.

user55340

re: annotations. If you could rewind the clock to 1.0 days and give them annotations... it would be very tempting to make static void main(String[] args) into something that was annotation based to say "when you invoke it as XYZ, invoke the method annotated with XYZ"

user55340

But you can't rewind it to 1.0 days... we've got static void main. But, in the EE world, things come in all over the place. There can't be "one" main method. So you've got web.xml

user55340

In the old days, you set up a xml tag that said "handle this request by this servlet" and "this servlet is this class" (two xml things... it was kind of a pain)

user55340

Now days, its annotations.

user55340

@WebServlet(name = "main", value = "/")
public class Main extends HttpServlet {

user55340

12:18 AM

@WebServlet(name = "Add", value = "/add")
public class Add extends HttpServlet {

user55340

And no complicated xml is needed for the basic things. You hit "/", and the container has looked up all the things with that annotation and does the initial routing for you.

user55340

This was one of the big things that the early frameworks did. "Everything that ends in '*.do' gets sent to struts, which then handles the routing"

user55340

If you want to look at one of those old samples of web.xml - docs.oracle.com/cd/E19857-01/819-6518/abxij/index.html

user20683

At what point does the transition from SE to EE occur?

user55340

It isn't a transition - they are different environments. EE builds on the SE codebase, but... its not a "transition"

user20683

12:21 AM

extension then?

user55340

Java early on "fractured" into three types. SE, ME, and EE. The ME was mobile edition.

user20683

I've got ME on my phone

user55340

You need to get a better phone...

user20683

@MichaelT I need to get a better life.

user55340

EE isn't a transition. Its a set of APIs and libraries (mostly in the javax.* namespace) that are used for large scale applications.

user20683

12:23 AM

ah

user55340

If you want to put something on a server, you could write a Java SE application that handles all its sockets and such. @Ampt was likely writing something like that with his web server application.

user20683

@MichaelT I suspect that's more painful.

user55340

Think of tring to write clustering software in any language... you don't want to. You want the OS to handle it.

user20683

@MichaelT It's like Controllers and Routing in Rails. Do I want to set all that up? nah.

user55340

Oh yea... EE... stay away from threads too. The container manages the threads, not your application. If you spawn a thread that the container doesn't know about, and... again, the application gets bumped to another server in the cluster, your threads didn't move with your application.

user20683

12:27 AM

@MichaelT It's fairly abnormal for me to spawn threads anyway. They aren't well supported in Python.

user55340

Personally, I like working in EE more than SE... because of all the restrictions. The design choices are 'easier'. Threads? Nope - send a request to another part of the application.

user55340

I used to work on a huge application that had two parts - an SE windowed application and an EE backend server.

user55340

The SE side had no end of thread messes. The EE backend... well, it couldn't... because you don't do it that way.

user20683

I'd definitely prefer a HTML/JS front end.

user20683

just package up my form data and punt it off to the server and sic the server on it

user55340

12:29 AM

So, with EE... you've got a container. JBoss or Tomcat or the like (note: Tomcat isn't a full EE server, TomEE is... for the most part this doesn't matter unless you taking that next step to clustering and deeper messaging)

user55340

Your application is deployed into the server, the server creates threads and handles the requests, dispatching them to the correct application and correct part of the application as need be.

user20683

because you followed the rules and set up all the right parts

user55340

Yep.

user20683

just like in Rails

user55340

Fun bit with most ruby app servers though... they're single threaded.

user20683

12:31 AM

which I did finish the initial "Here is an app that does nothing and is on Heroku"

user20683

hence the whole "has issues at scale"

user55340

Very much so.

user20683

There are certain things I like about ruby

user55340

And when you start getting into the 'well, we'll do a multithreaded server', they don't always play well with each other.

user20683

the ! and ? conventions for instance

user55340

12:32 AM

I'll give that its a nice convention... though makes some other of their syntax that they brought into it a bit messier because of other uses of ! and ?

user20683

@MichaelT I took a Parallel computing class. I wasn't terribly good at the coding part of it (my C was terrible, still's not so good). The theory mostly sunk in.

user55340

Btw, a simple EE app - github.com/shagie/TestingWithHsqldb

user55340

something I wrote to do some unit testing against an hsqldb database.

user20683

unit testing is something I need to do more of

Matt D

just remember there's multiple types of unit testing

I got into a discussion with my boss, i like blackbox testing. he likes whitebox testing. too me a while to realise that, we both thought each other was an idiot until it clicked. then it was like 'oh ok sure'

user20683

12:35 AM

@MattD unit, integration, functional, configuration (I think)

Matt D

blackbox/whitebox unit testing, there's another one i think

user55340

@MattD Those tend to be associated with people doing testing of the application and if they know the internals of how it works or not.

Matt D

yes. i prefer blackbox testing. but in some cases whitebox is useful. like when you're calling external services you dont "own" and cant look into their database to see if shit actually happened

user55340

Unit testing is testing a very specific piece of code.

user20683

def getThing:
return thing

user55340

12:38 AM

I wrote a class that implemented arbitrary fixed point math (so that 0.3 + 0.3 + 0.3 = 0.9 exactly)

user20683

assert thing is an instance of thing

user55340

(and I can do math)

user20683

@MichaelT I have a book on my shelf that describes the IEEE floating point standard in mind-numbing detail

user55340

Test to make sure that the scale of the answer when you do "0.3 + 0.3" remains 1.

user20683

I need to sell that thing

user55340

12:39 AM

@WorldEngineer The numbers were floating point.. they were a long and an integer. The integer indicated the power of 10^-n the long was to be multiplied by when done.

user55340

So 0.3 == 3,1. And 1.33 = 133,2.

user55340

That way all the math remains integer math - very important.

user20683

sorry misread

user20683

banks don't like floats

user55340

It was doing math on money... remember, I was in Point Of Sales before my current job.

user55340

12:41 AM

I was taking out all the Doubles in a particular section of code to get rid of the rounding errors.

user55340

all the operations were unit tested so that I could be sure that I was writing the code properly.

user55340

But the key thing - the unit testing was testing one operation and making sure the result was exactly as I expected.

Matt D

4:46 AM

ah floating point maths

i used to work in telco billing

lots of fun times with rounding errors

gnat

7:58 AM

@WorldEngineer members.shaw.ca/iksbloodoath/klingon_humour.htm

12.
Specifications are for the weak and timid!

11.
This machine is a piece of GAGH! I need dual Pentium
processors if I am to do battle with this code!

10.
You cannot really appreciate Dilbert unless you've read it in
the original Klingon.

9.
Indentation?! -- I will show you how to indent when I indent your
skull!

8.
What is this talk of 'release'? Klingons do not make software
'releases.' Our software 'escapes' leaving a bloody trail of designers
and quality assurance people in its wake.

7.

gaussblurinc

12:15 PM

Hello! Can anybody gives me advise about building service-oriented server infrastructure?
I want to build server, which support several steps of application maintenance. Git support / app building support / tests / error handling / developer's email feedback / Logs (of course) / {something else here}. Is it a good way to use services as units of this server?

user20683

'It has a ghost in it. Take it back.' 'No.'

Oded

Heisenbugs === ghosts?

maple_shaft

1:25 PM

God this is such a horrible noob question that I am too ashamed to ask it on the main site

In java, when I implement CompareTo, here is a scenario...

compareTo(Object A, Object B) ... a correct implementation must mean that given A and B the same score will always result

Compare A and C and the same score will also result every time

crap... I am not explaining this well...

Thomas Owens

Can you just autogenerate compareTo? It would be easiest. Be sure to autogenerate compareTo on any custom classes as well, all the way down the comparison tree. Then you don't need to think about it anymore.

maple_shaft

Lets say that my implementation of compareTo could return a number between 0 and 64 for a set of no more than 64 unique objects...

Thomas Owens

I guess it's only hashcode and equals that can be auto-generated. Not compareTo.

maple_shaft

Lets say that in my algorithm it would normally score B and C as 0, functionally equivalent for comparison

but

now there is a secondary scoring requirement for a tie breaker

Lets say B and C with the tiebreaker is now -2

err

I mean 2

but lets say A and B are also 2

what I don't want is Collections.sort to determine that A, B and C are all the same

Oded

You may need a ternary tiebreaker

Thomas Owens

1:34 PM

> It is strongly recommended, but not strictly required that (x.compareTo(y)==0) == (x.equals(y)). Generally speaking, any class that implements the Comparable interface and violates this condition should clearly indicate this fact. The recommended language is "Note: this class has a natural ordering that is inconsistent with equals."

maple_shaft

I want B and C to be scored relative to each other

I am not sure compareTo works that way

AHHH

Thomas Owens

I think that's applicable to you. Normally, there is a relationship between equals and compareTo. Collections should use compareTo for ordering, but equals for equality (like set insertion or searching).

maple_shaft

@ThomasOwens YES! That implies that equals and compareTo need not be consistent!

Thomas Owens

For something like the Integers, it makes sense that x.compareTo(y) == 0 means that x == y. But for sports teams, two teams could have the same record and tied for place, but not be the same team.

maple_shaft

though they recommend it

That implies that I can put logic in the algorithm that can make B and C have some crazy number that doesn't make sense when comparing B and C to any other number

Thomas Owens

1:36 PM

Yep.

It should be documented, since it's not usually obvious (depending on your domain).

maple_shaft

sure a sort operation might have to do a little bit of work to make sense of it, but it should work

@ThomasOwens ty! I will definitely document this class appropriately. I must have read that documentation 5 times and that never clicked in my head... apparently I need another cup of coffee

Thomas Owens

Oh. I'm not sure if you mean compare or compareTo, but they both have the same notes on usage.

maple_shaft

compareTo

Ya know... another possible way to do this would be to just multiply the primary scoring algorithm by 1000

secondary scoring algorithm will be in 100's

then as @Oded suggests if I need a ternary operation that can be in 10's

then it can work well with equals

user55340

3:06 PM

@WorldEngineer I still am tempted to add xkcd to the room feeds...

user55340

@gaussblurinc What are you trying to do? There's several bits there that you're describing... whats the big picture? What does the server / application do?

user41796

@MichaelT might as well add codelesscode while you're at it then. :-)

user55340

@GlenH7 Eventually, the XKCD gets posted here... Especially if its making fun of Haskell. Might as well automate it... the xkcd posting that is, not the haskell making fun automation.

user55340

Btw, I linked thecodelesscode.com/case/121 into the "two problems" answer...

user41796

@MichaelT I need to get an RSS aggregator going again. I looked around a bit when Google dropped Reader, but my Ludditism has held me back from finding another one to stick with. That or I've been too busy.

user55340

3:22 PM

@Oded Watch out with that === -- as I read it, that also implies type checking in javascript, and identity in php... never know what language the ghosts code in, but I suspect its postscript (who else would have written gostscript?)... and that's a nice, stack based language where I'm sure many bugs/ghosts can hide, waiting for the right rendering engine to be displayed on.

Oded

@MichaelT - that's a heisenbug right there then... different semantics depending on language ;)

user41796

So there is type consistency in this case

user55340

@Oded I recall my father learning to program postscript from the red book ( partners.adobe.com/public/developer/en/ps/PLRM.pdf )... in Microsoft Word on the Mac back in the day. He was creating a hand coded letterhead for UW Madison.

user55340

You set up a funky style that word read as "send this as raw postscript". The challenge was debugging it - modify, print page, modify, print page, modify, print page... there were no graphical post script rendering programs back then.

Oded

Heh - talk about a long feedback cycle, eh?

user41796

3:37 PM

Oh, the "good old days" ... ;-)

Oded

@GlenH7 with timesharing systems (punchcard days), I understand they would run the batches overnight, so you would only find out about a syntax error the next day ;)

I can't imagine what kinds of problems hanging chads caused though

user41796

Even though I was a wee lad, I remember my dad upgrading the RAM in the family's 8088. At some point in the process, all of us youngin's were banished from the room so he could concentrate and make sure he didn't bend any of the pins while inserting the chips.

user41796

@Oded bad enough that an entire lexicon was built around it.... :-) Never mind elections hanging in the balance.

Oded

I remember my dad bringing in punch cards.

user41796

At that point, mine was doing more sysad work than anything else. There were quite a few of the tape reels that he would have in the back of the car. Old school courier service FTW

Jimmy Hoffa

3:46 PM

This all just adds to the belief I have about second generation computer folk having a huge legup over others, though as computers normalize throughout society that advantage shrinks I think

user55340

@Oded you used a chadless punch card. catb.org/jargon/html/C/chad.html

Oded

@MichaelT surely, using termites would introduce bugs to the program?

user55340

The neat stuff about the punch card... its probably the most permanent memory storage that we've got for computers. The paper is heavy, unlikely to decay over time.

user55340

More durable than paper tape. Magnetic tape, drives, burnable CDs... these are all things that have decay rates in the range of years.

user41796

@MichaelT The biggest risks are a) having equipment that can whatever the media is along with b) having software that can handle / understand / utilize whatever is read.

user55340

3:52 PM

I recall some digital photographers early in the life of digital and burnable CDs lamenting that they lost 10 CDs (thats a LOT of pictures) that just... decayed. The burnable CDs use a dye in the plastic to different 0 from 1... and that dye faded, making everything 1. No way to recover the data.

user55340

@GlenH7 Punch cards are also "macro" enough that a human can read them by hand if necessary. Tedious, yes... but possible.

user41796

My understanding is that the media usually can last the system lifespan

user20683

@MichaelT FORTRAN as cuneiform

user41796

@MichaelT All of my experience is at the enterprise level where the media was more robust. But I do recall tales of the earliest CD-Rs and their lifespan issues. Regrettably, people often only found out after it was too late to do anything about it.

user55340

This was mid 2000s (2005 or so). "consumer" grade CDRs. The life span was sometimes measured in a few years. If you don't store the CDRs in an appropriately dark area, the breakdown of light on the dyes can be devastating.

user41796

3:56 PM

I have some old iomega zip drives that I should see if they are still readable. The problem is I don't think I have a system I can connect the drive to

user41796

So much for my copy of my thesis. :-)

user55340

@GlenH7 Possible, but kind of doubtful... they were high quality floppy drive at a very dense rate. The reason the old 5.25" disks are still readable is the very low bit density on them. It took a lot of degradation to erase a bit.

user55340

The Jazz Drive and similar platter based ones would likely be better, though their problem was that dust could get in more eaislly. (Sysdisk? Sydisk? something like that...)

user41796

Ugh. The jazz drive I was given was wretched. But it was kept in a dusty place, so that makes sense

user41796

I used to have a degausser, but gave it away on the last move. I'd like to be sure the zip disks can't be read before chucking them into e-cycling

user55340

4:02 PM

Jimmy Hoffa

@MichaelT I think you're right, that guy on Friday wasn't a troll, he's just really green and over-confident - he's since asked a grip more questions around SO and while he's got huge holes in his knowledge he enjoys telling people that he already knows things they try pointing out to him

user55340

The profile of a "about time to do a release" in github.

Jimmy Hoffa

@MichaelT "Branches? We don't have to show you no branches...we don't need no steenking branches!"

user41796

@JimmyHoffa I'd still call that a troll, imo

Jimmy Hoffa

@GlenH7 Just more unwitting than willful though

lots of "gimme teh codez"

user55340

4:03 PM

So... C2 was the first wiki. Some people were talking about virtual communities on there, but that spun off to its own wiki called 'meatball' - meatballwiki.org/wiki/WhatIsaTroll

user55340

> Trolls are of two kinds: Troll by intent which are made purposely and Troll by result which are not intended to troll in the first place.

user41796

@JimmyHoffa gimme gimme gimme!

Jimmy Hoffa

@GlenH7 You don't want mine, it's just variations on rm -rf with macro expanders to zygomorph it into what I please

user41796

/* works just fine IMO

user41796

Any problems I have promptly go away after running that one....

user55340

4:06 PM

Btw, @GlenH7 I've tossed a fair bit of things into vote to delete if you want to join in that part of cleanup... as we lack diamonds so far.

user41796

@MichaelT I think SE is purposefully toying us at this point. Just for sport. But I don't blame them as I'd do the same...

user41796

I had a busy weekend getting caught up from a family trip, so I slacked off this weekend on maintaining the site. And I need to catch up on my close reviews

Jimmy Hoffa

Family trip? Sounds fun

user41796

Wife graduated with her masters, so I took the kids to the graduation ceremony. Was up at Lied Lodge, which is a really pretty place especially if you like trees.

user55340

My niece (3.8 years old) is apparently asking when we're going to the Ren Fair again (the one down on the WI / IL border). I said "after the fireworks in summer"

user41796

4:09 PM

That's a good marking point - she'll remember the fireworks.

Jimmy Hoffa

@GlenH7 awesome! what's the paper?

user55340

The ren fair is the thing to do in July. August is blueberry picking.

user41796

@JimmyHoffa MFA in writing

user41796

@MichaelT I am a bit ashamed to admit I have never been to our local ren faire despite having wanted to go for years. Perhaps this is the year I should drag the kids out there.

user55340

The Bristol one is one of the best I've seen... and I've got to a number of them (2 different locations in Cali, 2x in Northern WI, Minnesota...). But part of that may also be my memories of that one.

user41796

4:18 PM

@MichaelT The memories can certainly bias you, yes. :-) The one out my way has been going on for decades and is supposed to be of fairly high caliber.

user55340

I got a bamboo flute there back when I was in 7th grade... the same flute maker is still in the same spot when I went there last year. Though now he's quite a bit older and his apprentices are the ones staffing the stall.

user55340

Bristol Renaissance Faire

Bristol Renaissance Faire ("Bristol") is a Renaissance fair held in a Renaissance-themed park in the Village of Bristol, Wisconsin. It recreates a visit of Queen Elizabeth I to the English port city of "Bristol" in the year 1574. The faire runs for the nine weekends from early July through Labor Day. History The Bristol Renaissance Faire was founded in 1972 by Richard Shapiro, and his wife Bonnie, as "King Richard's Faire". The event was a four-weekend fair and drew approximately 10,000 people. Initially the faire was generically historic, but through the 1970s and 1980s non-historical ...

user41796

That's awesome. I hope you told him you still have the flute you bought a few moons back.

user55340

Yep. I brought it with me once recently... his style of denoting that he made it has changed over time. He recognized that as a rather early one.

user41796

@MichaelT All of the folk I have known involved in that community would have been quite happy to see their handiwork on a repeat visit.

user55340

4:22 PM

(you know we're gonna make Thomas go "guys!" again...)

user41796

In the local scouting council, we had a guy who made a huge number of neckerchief slide blanks. He passed away a few years back and left a huge hole in the community.

user41796

But now I'm just adding fuel to the fire.

user55340

But yea... the ren fair... its something memorable.

user41796

I have a couple of bucket list items I have wanted to do this year, so I think I'm going to put that on the list.

user55340

Its fairly... substantial for a 3yo to remember wanting to do things from a year before, but that's on her list of things that is part of the summer.

user55340

4:27 PM

And this year my nephew (he'll be 2.5 in July) should have a better time too (he wanted to do things, but was scared to do them...)

user41796

I'd have to agree. Anything they're asking for like that is a good sign to do it again

user55340

Though an odd bit of today's kids, one of his favorite things to do is to look at the pictures on my parents' phones... especially the movies of my niece (his older sister) on the rides at the renfair.

user41796

@MichaelT I keep reminding myself that kids interact with tech in vastly different ways than what we're used to. And if we're smart, we won't judge that behavior but try to learn and understand how they see things.

user55340

They're immersed in it... its part of their life.

user41796

The relationship we have with tech is changing, and each generation has their own approach to it.

user55340

4:30 PM

►

user41796

Even the Gen Y's that I know - the older ones interact with tech in much different ways than the ones at the younger side of the spectrum. The up-and-coming generation treats it even more differently

user55340

Is it still Gen Y? Or are they Millenials now? (I get confused with the terms)

user55340

Wikipedia says...

user55340

Millennials

Millennials, or the Millennial Generation, also known as Generation Y, are the demographic cohort following Generation X. There are no precise dates when the generation starts and ends. Commentators use beginning birth years from the early 1980s to the early 2000s. Terminology Authors William Strauss and Neil Howe wrote about the Millennials in Generations: The History of America's Future, 1584 to 2069, and they released an entire book devoted to them, titled Millennials Rising: The Next Great Generation. Strauss and Howe are "widely credited with naming the Millennials" according to jo...

user41796

@MichaelT I call them Y. But I don't care too much about whatever the label may be.

user55340

4:39 PM

Just hearing the "millennials" on the news more than "Y"

user41796

I'm sure it's more PC or something. :-)

user55340

And the next batch is "Generation Z" until they get more of a name yet.

user41796

But PC still means personal computer to me

user55340

I think its more of "Gen X" wasn't descriptive enough when compared to the previous generation names.

user55340

Lost Generation, Greatest Generation, Silent Generation, Baby Boomers, Generation X (that name stuck), Generation Y, Generation Z

user41796

4:41 PM

Once again, the Boomers are screwing over Gen X. :-D

gnat

discovered a funny way to "indirectly cool down" our hot questions. I go to SO questions in the hot list and lemming-upvote stuff in there, causing them go up in the list. Somehow, I don't feel bad about this, given that per my observations, they somehow manage to keep their hot questions under control anyway (and this in turn may be the reason why SE doesn't give a shit about the damage done to smaller sites)...

@djechlin the difficulty is that no other site is SO. No other site has the number of active moderators to protect questions quickly, 20k+ users to delete bad answers, and the shear volume of other posts to cover up some of the damage on the front page. On smaller sites, we may be lucky if there's a mod on at that time, or a 20k+ user (much less three) and the front page moves at a glacial pace compared to SO. What works for SO doesn't necessarily work for the other, smaller sites. — MichaelT 2 days ago

user41796

"Oh no, you don't get a real generation name. But everyone after you? You betcha."

Jimmy Hoffa

@MichaelT "silent generation" -> Never heard of this?

user41796

@gnat that's hilarious

user55340

> The Silent Generation, also known as the "Lucky Few", were born 1925 through 1942. It includes those who were too young to join the service during World War II. It includes most of those who fought the Korean War and many during the Vietnam War.

user41796

4:42 PM

It's a bit passive-aggressive, but whatevas

gnat

@GlenH7 sweet revenge

Jimmy Hoffa

@gnat That's spectacular.

@MichaelT Interesting, no wonder I never heard of them... heh

user55340

@JimmyHoffa They're kind of lost between the "Greatest Generation" and "The Baby Boomers"

user55340

Btw, fun MSO post (on homework)

user55340

-17

Q: dear mods I need your help badly. I wish for my thread to be fully closed

Dear mods I had a thread as user2994348 and I wish for that thread to be fully deleted. Can a mod please do that?

discussion

user41796

4:52 PM

@MichaelT the comment thread is epic

user55340

@GlenH7 Yep... very amusing.

user55340

Oh, next in my "why certain questions don't make good questions" theme, I'm thinking "blog post citations"... the "I read this in a blog post, why is it wrong?"

user41796

@MichaelT Oy that's a can o' worms if there ever was one

user41796

Is it wrong to place a comment on that SO question linking back to the MSO request?

user41796

"Relevant meta question ..."

user55340

4:59 PM

@GlenH7 Not entirely off topic... and also would serve as a pointer for the prof when she finds that text of the user trying to cover up.

user41796

done

user41796

My hope is that I have added a little justice to the world

gnat

10Kers - @GlenH7, @MichaelT - what do you think about this? meta.programmers.stackexchange.com/a/5542/31260

user41796

@gnat sneaky espionage?

user41796

reading now, btw.

user55340

5:22 PM

(that two problems regex answer is still giving some nice rep even after its fallen off the hot questions)

user55340

@maple_shaft btw, thank you for the protection on that php question... just noticed it climbed up on the hot questions somehow.

user55340

It probably would have been a CW mess by now if you didn't clean it up. Thank you.

user55340

@GlenH7 The blog citation thing was inspired by that "when is code bloat good" one which was referencing another P.SE answer that cited a blog that was about the "good enough" software is what sells.

user55340

Blogs are about drawing readership (and discussions to maintain that on older posts)... to draw the readership they sometimes make outrageous claims like how great lisp is or other silly things.

user41796

@MichaelT and oh can they attract lemmings....

user55340

5:36 PM

The "here's a citation, prove it wrong" or worse "here are two warring citations, which is right?"

user41796

@GlenH7 While usually I'm pro linking to a related meta discussion I don't want the OP to get into extra trouble for trying to hide cheating on their work — Richard Tingle 9 mins ago

user55340

The only times I've seen it work is twice... the famous Alan Kay answer and once on P.SE when someone quoted someone's twitter comment and that person happened to show up to explain it.

user41796

There is a little less justice. Richard's comment was up-voted so I deleted the link back to MSO

user41796

@MichaelT generally doesn't work out that well, that's true

user55340

Incidentally, I can't find that warring blog posts question of Jeff Atwood vs the "10 years to master a skill"

user41796

5:41 PM

@MichaelT mod deleted maybe? I don't recall it rolling through the 10k delete queue

user55340

It was awhile back... could have been deleted prior to us being 10ks...

user41796

I was a bit surprised to see I'm ~800 away from 20k. C'mon faster deletes! :-)

user55340

@GlenH7 1.8k here.

user41796

I am not going to have a problem burning through delete votes today...

user41796

@MichaelT I ought to jump on a few collider questions when they're still early

user55340

5:50 PM

php, php, badly thought out theory question (I do hope @amon gives an answer to that one...)

user55340

There's that unit testingish one...

amon

@MichaelT you summoned me? What question are you referring to?

user55340

@amon The lexing one. Could do with a good theory answer.

user55340

You've got a comment in there already.

user41796

What's the delete vote required rule? There's a question needing deletion, but says it needs 2 more votes. I think some judicious down votes will push it to just one delete vote. programmers.stackexchange.com/q/57360/53019

user55340

5:52 PM

3 is the minimum vote to delete.

amon

@MichaelT ah yes, I'll take a deeper look

user55340

@GlenH7 One needed now.

user41796

@MichaelT none needed now. :-) Thanks for the down votes.

user41796

I'm on the fence with this one. programmers.stackexchange.com/q/213306/53019 Essentially, it's asking if stealing is justifiable. But we also had a very well voted question + answers asking if it's okay to take the source code with you when you leave a job.

user55340

@GlenH7 then skip it... mull it over. There's no rush on any of them.

user41796

5:57 PM

I'm wondering if it could be saved with a heroic edit...

user41796

As phrased, there is no disputing the answer - it's unethical and potentially illegal

user41796

But you could ask about what conditions would allow for that so it would be ethical. ie. CC licensed sites or something

user41796

But that edit would kind of invalidate the top voted answer

user55340

6:11 PM

@GlenH7 Thing for me was to look at the close reason:

user55340

This question appears to be off-topic because it is about legality and this site is not a legal advice site; and this question is also about ethics, which can be highly subjective. — FrustratedWithFormsDesigner Oct 3 '13 at 19:54

user55340

Its the type of situation where a faster close might have allowed it to be redirected before the OP got an answer that went off topic.

user41796

@MichaelT It's the difference between "should I steal?" vs. "do you steal?"

user41796

Neither are a good fit for the site....

user55340

6:38 PM

See review edit adding links and generally fixing up the material... scroll down... Yep. Peter. Hit approve. I know I'll be in trouble with an audit if his name ever shows up as the fake name.

Jimmy Hoffa

@MichaelT I just approved one of these -> 'Peter' ? I didn't notice the name, has someone in particular been doing a bunch of these?

programmers.stackexchange.com/review/suggested-edits/49253

Yeah, Peter. I see now.

user55340

@JimmyHoffa Peter is a... prolific adder of links and fixer of grammar.

user55340

@JimmyHoffa programmers.stackexchange.com/users/591/…

user55340

Or if you want to see big numbers... stackoverflow.com/users/63550/…

Jimmy Hoffa

holy shit. he has 717 rep and 566 revisions

Oh geez.

user55340

6:45 PM

#3 for most edits on SO stackoverflow.com/users?tab=Editors&filter=all

Jimmy Hoffa

@MichaelT the fact that he's #3 speaks to the obsessionist insanity that makes SO both great, and avoidable.

user55340

@JimmyHoffa The 301 suggestions is a better one to work from when comparing rep. The 265 other revisions are likely in CW posts. 301 suggestions = 602 rep out of 717 for 115 rep from Q&A.

user41796

The nice part about Peter's edits is they are (almost) always demonstrably better. So it's a no-brainer to approve and move on. I'll feel bad when he rep-caps on number of edits (no more than 1k rep from edits).

Jimmy Hoffa

@MichaelT 15 actually, 100 surely is bonus

user55340

Makes sense.

user55340

6:53 PM

@GlenH7 Could do a "give big bounties to his existing Q&A so that he can get to 2k rep...."

user41796

@MichaelT He may also be happily following a lurk & learn MO. So edits are just what he does and may not mind accruing the rep from them while he lurks.

user41796

But I certainly wouldn't rule out some extra rewards for helping maintain the site like that.

Jimmy Hoffa

@GlenH7 I suspect 566 revisions -> he's lurked more than long enough. He probably just likes correcting english (not his native language)

user55340

From the grapevine known as "google hangouts with people I used to work with"... so two of my former co-workers were interviewing last friday. Today the team lead (who is late 60's) announced his retirement.

Jimmy Hoffa

Heh soon that place is going to find itself full of software without a single programmer eh? You gotta wonder what happens in a place like that

user41796

7:03 PM

@MichaelT oh wow! more openings. I should get a call back from them any day now!

user55340

Oh, did you see the news article about @Ampt 's future employer?

user55340

zdnet.com/…

user41796

thanks for the link; I hadn't.

user41796

Honestly, that's the type of stuff they can knock out of the park like no one else's business. They've got a preferred scaffold they like to follow and they can churn things out against it.

Jimmy Hoffa

Shit. Accenture? UGH

user55340

7:05 PM

(I'd honestly say thats a good thing for Ampt... being able to put down such a high profile project on the resume, even if you just fetch coffee...)

Jimmy Hoffa

@GlenH7 Seriously? You think they're going to put together something that's not bug-filled git-r-done to get it out of their shop?

user55340

@JimmyHoffa its more of the thing that they know red tape, they've done big projects and they don't have a silicon valley mentality.

user41796

@MichaelT I gotta agree. That's a bread & butter project for them. So, yeah, @JimmyHoffa I think that corp will do just fine with that project.

user55340

Google / Oracle / whatever out on the left coast, they want to build something and toss it out there and move on to the next shiny. Oracle less so than Google, but they're tightly bound to their own dog food.

user41796

It plays to many of their strengths. The site doesn't need any super-dooper innovative type stuff. It just needs to get done. And that firm can knock 'em out in that case.

Jimmy Hoffa

7:07 PM

@MichaelT I realize they know red tape, but still.. they're going to fix it in one year? call me crazy but that's either going to get extended and extended and extended, or it'll be "Look we did it!" in one year which is going to be a nice paycheck for them and less functional than before

user55340

I suspect that much of the problem isn't the technical, but the social.

user41796

@JimmyHoffa oh, it will turn into a never ending project, no doubt there. But that was going to happen no matter what. It's why CGI was so giddy when they landed the contract the 1st time it went out.

Jimmy Hoffa

@GlenH7 If that's the case I'd be totally happy with Accenture, but I suspect because of the pressure on this whole thing the government is going to absolutely hate such an idea

Think of the political fodder -> 2 years later "Accenture is still working on fixing ObamaCare, let's fix it with repeal now!"

user55340

(whee! validation of my answer on regex)

user55340

@KonradRudolph: Like any tool, quotes can be used for both good and ill. I use regex almost daily. Very powerful. Has gotten me out of many jams. I encourage others to learn it. MichaelT's answer helped me to see regex not so much as a "just another library call", but a language unto itself... and very different from many of the languages that "host" it. I also acknowledge regex is just one step above line noise in appearance (write once, read never :-), thus I can understand people's reluctance to invest the effort to learn it. Unfortunate, but part of life. — JS. 2 mins ago

user41796

7:11 PM

@JimmyHoffa I'm more hopeful (obviously). They've got the resources and I'm sure they have done similar projects before. They have a structure in place to get a more robust system in place in a short period of time. So I'm wiling to take a wait-and-see approach rather than betting on their failure.

user55340

@JimmyHoffa Occasionally touching on projects with other groups in the federal level... there are many worse projects that are even recognized as "contractor welfare" type things.

Jimmy Hoffa

@GlenH7 Fair enough. I'm just thrown by the things I hear of the current system which makes it sound like they didn't do simple so Accenture's either going to be stuck with this behemoth monster or a rewrite and neither sounds like a year gig. If they get the continual and don't just have unbelievable amount of pressure put on their contract then I think you're right, they'll knock it out.

user41796

2

A: Can you apply the same lexer rules to all programming languages?

No you can't. Here is a nonsense snippet of Lisp: (*foo-bar* 'baz 14) What are the tokens in this snippet? The answer is something like LPAREN SYMBOL "*foo-bar*" QUOTE SYMBOL "baz" INTEGER "14" RPAREN Lisps have very liberal rules what can be inside an identifer: hyphens -, asterisks * an...

user55340

@JimmyHoffa consider - it needs to be fixed. A tech company isn't the one to go in and try to do maintenance work on it (they might be able to address some of the hardware... but you need a contractor to come in and actually fix it)

user55340

@amon superb job!

user41796

7:15 PM

@JimmyHoffa They'll just throw out all the sh*t that was previously done. They're getting good at salvaging custom projects by re-using the business logic and trashing the rest. But the popular press won't be told it's going to be a complete rewrite as that's more fodder....

user41796

@MichaelT yeah, I thought that needed some up-vote love.

Jimmy Hoffa

@GlenH7 Yeah I'm sure nobody will say it's going to be a rewrite, but if it is -> 1 year? That's going to be rough.

user55340

@JimmyHoffa you're going to need a big company that can throw the manpower at it and has dealt with 100-1000 sized developer projects before and is aware of what that entails and how to make it work.

Jimmy Hoffa

@MichaelT Perhaps. But company's like that move even slower than the ones used to doing 5-15 man projects.

user41796

@JimmyHoffa I didn't look too closely at the numbers, but it's a limelight project for them. I wouldn't be surprised if they took a lower projected profit margin just to secure the contract.

user55340

7:18 PM

Very true. But the thing is, companies that have never done a 100 man project before don't know what it takes to scale up from there.

user41796

@JimmyHoffa You would be surprised. Their strength is their reliance on the scaffold they use. They can break really complex projects down into manageable chunks very well.

user55340

I've seen one 100+ man project before... that was implementing a new CRM at Netapp. It took longer than expected, but thats also because of how the thing worked and it was a "do all these things."

user41796

And mind you, I used to work for a competitor of theirs. So for me to sing their praises... :-)

user55340

This is a "fix all these things", the requirements are rather set down at this point, and thats the worst thing (the requirements gathering).

user55340

You aren't going to have "oh, and we want a pony" in the project now.

user55340

7:20 PM

They need a 100+ man team to work on it, and I can't think of too many companies that fit that role with available staff.

user55340

And you really don't want to get to the slightly smaller "ok, lets drop all future contracts and start hiring independents to bulk up the pool"

Jimmy Hoffa

@MichaelT Company's like Accenture often times do the ladder part their for most of their projects has been my understanding -> "We got a contract! Now let's hire some people to put on it."

user41796

@JimmyHoffa nah, they run a deeper bench than that. And with a 1 year contract, they won't necessarily staff up for it.

user41796

multi-year contract, sure. But for a shorter term contract, not as much.

user55340

At my former contractor, I've seen the 'preferred contractor' get tapped out eaislly (5-6 people) on one contract, and then hiring independents for the other one offs that come out... and thats not a good thing (they have their own timelines and when the requirements gathering goes on too long, they have their pre-existing commitments)

Jimmy Hoffa

7:24 PM

@GlenH7 It's not about their bench, it's a business model, perhaps I'm mistaking Accenture's then. I assume they're like many other large scale consulting/contracting places that just maintain a staff of business/marketing people and some high level tech folks to help them make the sales bidding contracts, and as soon as they get a contract they staff it up with whoever, hand it over to their collections department and forget about it

There's lots of company's like that which maintain basically no technical staff that's not on-job and only hire when the contract is signed picking up anybody who they can pass-off as able. These are often the same places who are very handy with the H1B process...

user41796

@JimmyHoffa To some extent that's true; my perception is that it's not quite as you're painting it though. Especially on a limelight project where they know they are going to get national media attention for the project. I'm almost positive they were very careful in how they structured the contract and they set themselves up for a knock-it-out-of-the-park type win so they could use it for future marketing collateral.

user41796

I think they maintain a deeper technical bench than you realize. My understanding is that attrition tends to be their limiting factor with maintaining staff. It used to be (and may still be) that a lot of folk bailed after making their 3rd promotion because they went from hourly to salary and they were put under sales pressure.

user41796

I would hope they have created more of a technical ladder since I interviewed with them, but their model has worked well for quite some time.

Jimmy Hoffa

@GlenH7 Aye, this is a somewhat different model then; the model I refer to is reliant solely on marketing talent to keep finding contracts as all past jobs were messes unless really lucky, you refer to a model which relies on a track record of good quality because otherwise they wouldn't retain the technical people

user55340

@JimmyHoffa Hmm... a scatter plot of "sales skill vs technical skill" would be something to see.

user41796

7:40 PM

apropos of nothing - how does Google make money off of Google Voice? I can't see where they are charging for the service.

user55340

@GlenH7 Information.

Jimmy Hoffa

@GlenH7 They charge

user55340

Remeber they are selling advertisements. The more they know about each person who uses their service, the more they can charge for each advert. Then there's also the enterprise solutions.

user41796

@JimmyHoffa they charge for domestic calls?

Jimmy Hoffa

@MichaelT they also charge for the service

I don't remember the exact details

but I setup an account recently and saw lots of options for charging etc

There's some amount that's free but then there's an amount of the features that cost

user55340

7:43 PM

support.google.com/voice/answer/…

user55340

> You can sign up for Google Voice and use many of our features for free including making calls to the US and Canada. The section below explains how using Google Voice on your cell phone may impact your cell phone bill.

Jimmy Hoffa

@MichaelT international costs money, but then they also have voice mail type features and forwarding stuff etc

I think you have to pay for certain amounts of that stuff.. I'll login and look again

user55340

Yep. They describe the way it works... that link is probably the one you want to start from.

user55340

Btw, should we should chip in to get Ampt a poster or something so he can enjoy his work? I was thinking... despair.com/consulting.html

Jimmy Hoffa

@MichaelT That's a little too motivational for someone as green as him. Now is the time to strike, break his spirit now and he'll never suffer motivation-pangs for the rest of his career

perhaps

user55340

7:48 PM

@JimmyHoffa If he gets on to testing HC.gov project - despair.com/government.html

Jimmy Hoffa

or...

user55340

There's also despair.com/achievement.html

user55340

and despair.com/acquisition.html

user41796

I thought he was getting the equivalent of a back-office gig anyway

user41796

and I'll play with google voice a bit before I request to have my number ported over

user41796

7:53 PM

speaking of which, I wonder if I should create a separate email account for the google voice account

Jimmy Hoffa

@GlenH7 He smells like a back office dev to me... don't recall what the particulars were for the gig though

user41796

I remember warning him about the travel but he said this one was specifically not a traveling gig.

Jimmy Hoffa

@MichaelT ...did you notice in that comic the alterations in 'card' spelling throughout?

...or the fact that the comic was created by

Guy L. Steele, Jr.

Guy Lewis Steele Jr. , also known as "The Great Quux", and GLS , is an American computer scientist who has played an important role in designing and documenting several computer programming languages. Biography Steele was born in Missouri and graduated from the Boston Latin School in 1972. He received a BA in applied mathematics from Harvard (1975) and an MS and Ph.D. from MIT in Computer Science (1977, 1980). He then worked as an assistant professor of computer science at Carnegie Mellon University and a compiler implementer at Tartan Laboratories. Then he joined the supercomputer comp...

user41796

@JimmyHoffa - did you set up an analog to VOIP adapter to use with your google voice number?

Transcript for