The Whiteboard - 2015-11-23 (page 2 of 3)

« first day (1907 days earlier) ← previous day next day → last day (3087 days later) »

Lightness Races in Orbit

5:00 PM

@MichaelT heh

@JimmyHoffa yeah but more than two VMs and RAM contention becomes a real bitch

Podcast #69 - It's Too Rainy For A Parade

Abby T. Miller on November 23, 2015

Welcome to Stack Exchange Podcast episode #69, brought to you by The Lake Erie Soda Water Company. Your host is Joel Spolsky, joined by First Deputy of Community And So Forth Jay Hanlon and Lord High King of Nerds David Fullerton. Fortunately, the beer arrived shortly after the podcast began, so this one should be pretty good.

Lightness Races in Orbit

where two is however many for your system

and if your VMs are not running a lightweight OS designed for the task...

Depends on what kind of VMs you're running - a linux VM rarely needs more than a gig or two of ram, depending on what you're hosting on it.

It's all about putting VMs in their place. Need to test something in an OS that's not your host OS occasionally, or otherwise infrequently use some specific application where there's no good version for your OS? Done. If you need to do something constantly that relies on an application your host OS doesn't have, change your host OS or dual boot etc.

to just have a linux environment for compatbility testing, 512MB is more than enough

5:04 PM

@MichaelT are you referring specifically to this MSO question or did other stuff happen?

user55340

40

A: Stack Overflow fatigue - has your usage / motivation dropped off?

My use of SO has decreased dramatically since the signal-to-noise ratio went through the roof floor. A lot of that is down to the fastest gun in the west problem, because when I answer I generally try to to be as verbose and explanatory as possible, which generally means my answer ends up getting...

user55340

157

A: Should trivial re-occurring questions really be answered?

A question like this would have been closed in a heartbeat in the olden days. But not anymore, it is hard to close it today. There's history behind this that's relevant. The two site founders, Jeff Atwood and Joel Spolsky, had very different ideas what kind of questions should be on topic for ...

cool, that's exactly what I was reading

user55340

The second one gives the history.

user55340

The thing to consider is how P.SE has moved further in the Atwoodian direction.

5:07 PM

which is the ultimate irony

Sorry, I should have clarified, in terms of %

wow, even Jon Skeet said "I find considerably fewer good questions to answer than I used to"

user55340

@Ixrec Its a matter of finding them. And it can be hard.

user41796

Most large companies rely upon the concept of a center-point within the salary bands

user41796

And they want to get you to that center point after n years. So they'll likely bring you in a few years shy of hitting the center

user41796

5:11 PM

@Ampt How many years do they have with the firm?

@GlenH7 they were in my incoming class.

@MichaelT I've probably mentioned this before, but the best thing about this close vote experiment is that now it's possible to find the good questions on this site again

3

@GlenH7 no clue on the "Standard" %

I was only told my %

user41796

Well, they'll never tell you what the standard rate is. You have to ask around and see what others got. :-)

Doh. I should have known that

user41796

5:13 PM

You also have to look at the types of projects your peers were on. The more they bleed for the firm, the more they get paid.

user41796

And IIRC, you had a rough start to your billable career

@Ampt here's the valuable takeaway you're supposed to get from this; one many of us have had to take away: Don't ever find out what your coworkers make. Unless you're in some kind of union or otherwise have leverage, you'll only wish you didn't know. No matter which way the bent, you'll wish you didn't know.

user41796

There's quite a bit of truth in that

yeah. You're probably right

I feel like I agree with that even though I have no idea what anyone else makes

user41796

5:15 PM

@Ixrec It's a no win situation.

user41796

@Ampt Getting a promotion that soon in is a good sign in my book. Usually it's 2 - 3 years before promotion.

user55340

I know what everyone I work with makes, because its the law. Seriously. The rates are set in the law.

user15026

Upstairs people broke the Internet. I hear them up there but they didn't answer my phone call to ask them why it is broken.

@Ixrec I found out incidentally once; I learned my lesson. I've actively tried to not find out those things ever again. It just makes you irritated over something you can do nothing about. If you're in an organized workforce then there is something you can do, but otherwise, you'll just feel better not knowing.

user55340

@AshleyNunn You're here, its not completely broken.

user41796

5:16 PM

:25634301 Hmmm, maybe they've change the rules up a bit then

user41796

So the promotion becomes a way of flagging that you've cleared your probationary period

user15026

@MichaelT I am here because I have unlimited phone data.

user15026

But my desktop says nope no Internet.

@MichaelT does that mean there's never a nagging "why do I make less/more than that guy" feeling?

I may just need to re-evaluate my expectations.

user55340

5:17 PM

@AshleyNunn Unlimited data is a nice thing.

user41796

@Ampt This. :-)

user41796

With mega-corps, it's very much a case of "take it or leave it."

user55340

@Ixrec yep. The only variations are the contractors... and what they are paid may also vary between them (as in the company takes different cuts and different companies take different cuts). But all permanent employees you know exactly what they make.

@JimmyHoffa enjoying the current project very much. It ends in March. There may be a re-evaluation closer to that date.

@JimmyHoffa oh you better believe that's the plan

@GlenH7 and it's important and valuable to recognize that is a choice, not a threat. Genuinely.

user41796

5:20 PM

@Ampt There's a lot to be said for job satisfaction too

user41796

@JimmyHoffa That's actually a good point that many forget

can't get a much better answer than that.

@GlenH7 absolutely, but I may be moving back closer to home to buy a house at that point

we'll see

user55340

The other bit with that is there is really no negotiating your wages in the public sector. Or its really hard unless you have specific super skills. And being an exempt employee means that you won't rake in the OT pay like bus drivers and prison guards when there is a staffing shortage (it also means they can't force you to work OT).

user41796

Savez teh cash and burnz down teh debtz

@GlenH7 yeah, I mean, really it's the best way to get ahead in our industry - unfortunate to say but "leave it" is a pretty high-reward gamble that is how many of us have constructed our careers. In-house promotions are typically a joke compared to out-house promotions.

5:23 PM

@GlenH7 doing that too.

user55340

Public sector - its the other way around. You can easily get an in house promotion with a pay raise...

@JimmyHoffa ugh this is so true it hurts.

user55340

Public sector is like bizarro world.

nothing like getting bent over for showing a little loyalty

@MichaelT truly. It's like I've said before, the variance in our industry is staggering..

user55340

5:25 PM

(several years ago, when the higher ups did stupid things. a good chunk of the IT department was able to transfer to other Departments (notice big d and little d there) within a month... most of them getting pay raises with the move up too... but technically staying in the same "company" - public sector for the state)

user55340

It was "hey, we're being stupid here" and one server admin when to DOT, another went to HFS (Health and Family Services) and another went to DOA... and that was an easy internal transfer.

user41796

Very cool

user55340

@enderland SDET tends to be more... accepting and not as focused on CS concepts. Its a 'get stuff done' with the recognition everyone uses different tools.

so that's promising

user55340

Employer^^ pretended they had SDET. But they didn't. The extent of their SDET SDE part was "I can write an automated test in one testing application". And the "T" part was "we follow the tests the developers write and say if anything didn't match." It was... yea. Not great.

5:33 PM

@MichaelT the majority of the industry pretends to have SDET... the biggest issue comes down to SE folk don't want to manage manual QA people, so they don't take the management role to hire SDETs. It basically has to be a new from scratch place with the foresight to hire an SE manager for the QA team otherwise you'll get manual QA folk in and SDETs will never make it in..

I have a hard time imagining how anyone can be "just a tester"; IME refactoring code to make more aspects of it testable is at least as hard a problem as getting the code to work properly in the first place

@Ixrec you haven't worked anywhere with 10-20 testers in the QA dept who spend 100% of their time writing step-by-step instructions on which buttons to click and text field values to use called "test cases" (which they claim is proof of their test "engineering" prowess) and executing said test cases by clicking buttons and filling out text fields.

it's extremely prevalent

oh manual testers

@JimmyHoffa this guy also had a long blog post about his attitude about testing/QA he wrote recently too, was a helpful insight

@enderland industry wide it's an absolute mess. If that place is sincere about proper test automation then you could definitely write your ticket with a job like that.

5:38 PM

speaking of test automation

@JimmyHoffa yeah. the vibe I got was, "we'll throw a problem at you and you figure out what you want to do to solve it"

or even a "start finding problems to solve and do it"

we recently moved our team's main code repo onto the internal github instance

and now we have a webhook that automatically runs the tests on every pull request

WHICH IS AWESOME

would be nice if it didn't take >2 minutes to run them all, but oh well

>2 minutes? that's a heck of a lot faster than ours were...

our coverage is kinda crap though

but, it's probably possible to break this app into multiple repos and speed things up that way

ofc, not gonna suggest that until people start complaining about too many tests taking forever to run

whenever I figure out how to make all the UI-related logic testable maybe we'll finally have that problem

I wonder if it'd be possible to make the webhook smart enough to notice which module was most changed and run those tests in a dedicated job to get a super-fast result

probably not worth the effort just yet

@Ixrec HAH! 2 minutes?? You are brainded. I have never seen someone get their CI that fast on any actual software... People just don't care and management doesn't know why it would matter for a shit

5:49 PM

@JimmyHoffa "brainded"?

This reminds me of a question I don't think is scoped right, but relates to... "how do you manage the scope of individual repositories within a very large code base?"

I think that question depends largely on what tooling is connected to all the repos

@Ixrec phonetically spelled and you still couldn't read it.. man, how are you even here??

feels somewhat like a circlejerk of, "make a single big repo!" to "this is unmanageable and takes forever" to "how can we manage linked versions" to "screw this, who cares"

e.g., I assume most CI stuff associates a "build" with one repo

so the library that X unrelated microservices all share gets its own repo

etc

5:51 PM

@Ixrec yeah, but the problems come when a repo relies on another repo

I do think we over-forked our backend code, because a lot of non-trivial changes there seem to involve 3-4 repos, thus 3-4 code reviews and possibly more than one new deployable to roll out

@JimmyHoffa oh, braindead, gotcha. I was reading it closer to "braindid" to which the closest real word was "branded" which made no sense.

also I'm still unsure whether you're saying 2 minutes is good or bad

user41796

@Ixrec It's 20-30 minutes to run our automated full build + tests.

@GlenH7 what languages? approx LOC?

I guess we're cheating a bit because you don't need to "build" Javascript

user41796

So they suck the bejeezus out of things

(minification and stuff is done automagically as part of the publishing process)

5:59 PM

@Ixrec I've had phone-home-licensing tools in build processes multiple times in the past.. that is a thing sometimes. Custom obfuscator at one job, InstallShield's installer builder will do this, it's not an uncommon thing.

come to think of it, our ugliest services do take 20 minutes to an hour to build, being 10s of kLOCs of C++

all of our backend tests are on the shared libs though, so those usually take a few minutes tops to build and run; all the really hairy dependencies are pulled in only for building the full service executables

we should probably have some integration test suites for the full services... but one thing at a time

6:16 PM

@GlenH7 @AshleyNunn @WorldEngineer @enderland @ThomasOwens Surely you each have at least 1 recipe you can contribute food, delectable, or drink. Can't just be my personal recipe book, that's no fun.

user41796

Surprisingly, I very rarely mix drinks anymore

user41796

even my martinis are just a dash of the cordial along with the vodka of choice

Lightness Races in Orbit

Can’t get over when Microsoft tried to define “podcast” as “a portmanteau of ‘portable on-demand’ and ‘broadcast’” http://www.theguardian.com/technology/2015/nov/23/former-bitcoin-companies-cryptocurrency

Tweeted by alexhern on November 23, 2015 at 1:47 PM

@GlenH7 I go back and forth. I like variety; I been drinking good booze lately, but I'll do random beer for a while then random mixed drinks for a while, I'm all over the place. All the same, food recipes are more than welcome too. Like I said given the season I'm going to toss my wife's cranberry tart recipe up when I can weasel it out of her...'tis awesome.

@LightnessRacesinOrbit ...the fuck...

@MichaelT any thoughts on the comestibles section initial partitioning?

Lightness Races in Orbit

@JimmyHoffa I even checked that iPod doesn't stand in any way for "i-Portable-on-demand". It doesn't.

6:28 PM

@JimmyHoffa It's like the US and east berlin during the cold war...

Lightness Races in Orbit

(It's actually inspired by 2001, of all things)

@Ampt because that made everything much clearer?

it just occurred to me, the K&R was written by A powerful king.

Lightness Races in Orbit

bbc.co.uk/news/science-environment-20772246 listen to the second one. chills.

6:47 PM

@LightnessRacesinOrbit its funny that a related article shows the V for Vendetta person, lol

Lightness Races in Orbit

@enderland is it?

oh cos "vendetta"?

user image

Lightness Races in Orbit

@enderland yeah I know what it is I was asking why it's funny!

@LightnessRacesinOrbit you don't find that image right after an article talking about how companies are trying to reinvent history amusing? :P

@gnat just fyi, Lilienthal is often in the Workplace chat too if you want a convo there

Lightness Races in Orbit

7:05 PM

@enderland IDGI

You might try submitting this question to Programmer's Exchange: programmers.stackexchange.com — Ira Baxter 19 secs ago

@IraBaxter that is a recommendation which will lead to this question being closed. Please avoid suggesting other sites unless you are familiar with their scope. Resource requests are explicitly off topic on programmers.stackexchange.com — enderland 45 secs ago

This really isn't the right place to ask this kind of question. If you can be more specific you can try programmers.stackexchange.com but stackoverflow is for specific programming problems, this is too broad or primarily opinion based. — Ron Beyer 18 secs ago

7:22 PM

we're totally not ganging up on that guy

7:43 PM

sunufuhbich I have to buy a coffee percolator because I've never had a cup from one before

user41796

@JimmyHoffa WAT?

aero press and percolator are the 2 cuppas I've never had

user41796

skip the percolator

user41796

seriously

? no good?

@GlenH7 I associated percolator with drip coffee, but it's completely different apparently

user41796

7:45 PM

Or just wait for your favorite local church group to have a morning event and have some then

@GlenH7 O_O the hell you say? you're trying to get me killed aren't you!

user41796

Just to be clear, you mean something like this, right:

user41796

user image

user41796

Much smaller and nicer now

@GlenH7 I don't know? I always thought "percolator" just meant drip coffee

user41796

7:47 PM

No, no. Two different styles

pourover might be what you are thinking of

user41796

very similar in how the hot water is delivered to the grounds though

user41796

A percolator has a heating element at the base. The hot water rises up a tube in the center and then spreads over the coffee grounds

@GlenH7 but the percolator continually brews I understand...so I could just let it go for like an hour and end up with every oz of coffee extracted thoroughly, straight into my belly

user41796

And that keeps repeating until you turn the heat source off. Tends to be quite bitter

user41796

7:48 PM

@JimmyHoffa bitter bitter bitter

@GlenH7 is the bitterness perhaps just a result of the beans used?

@GlenH7 hey some people like that crap, it's basically starbucks coffee :P

user41796

And the main diff from drip coffee is the water passes through the grounds only once

user41796

@JimmyHoffa Nope, I've tried them with low grade and high quality beans. Same result.

user41796

Slightly better with the high quality stuff, but still on the bitter side

user41796

7:49 PM

Not a big deal if you drink it with tons of 1/2-n-1/2 and sugar. Is a big deal if you drink it black.

user41796

I'd put my money in the aero-press thingy and / or a pourover

user41796

The real advantage of the percolator is that it makes a lot of tolerably drinkable coffee in short order

a good pourover is my favorite type of coffee, no heating plate or anything to really make the coffee taste burned (unless you dump steam through the beans, I guess)

user41796

@enderland Have you seen this one yet

@GlenH7 I almost exclusively use a cheap plastic one I got at Ace Hardware

A friend of mine has a chemex, which I'd want if I made more than 1 cup at a time but my wife doesn't like coffee and I normally don't need more than a cup/thermos at once

user41796

7:54 PM

Cool. I'm considering one of those as a present to myself

I think they are bigger normally than a single cup

this is basically what I have - bedbathandbeyond.com/store/product/…

user41796

@enderland Does that have a little trap to hold the water in place until you put it over a cup?

@GlenH7 no... I just put it on my cup initially

or thermos

user41796

@Ixrec I know, right? They steal one of this room's main topics and turned it into a whole SE site...

7:58 PM

@JimmyHoffa The US refused to acknowledge east germany as a real country. It was a big deal for the east germans who wanted to be recognized. It's a lot like the Israel/Palestine thing today.

If the US were to come out and say Palestine was a country, that would instantly grant them a ton of protections and rights as a country, which Israel doesn't want.

Its a lot easier to squash a group of people when you don't officially recognize them as a group

@GlenH7 Those bastards....

or you officially recognize them as "insurgents" or a "terrorist group"

@Ixrec This as well, but I think that's more of a recent thing.

true

@Ampt well to be fair russia didn't exactly give them much option for being an indepdendent country...

ALL I WANTED TO DO WAS MAKE A JOKE AND NOW YOU'VE ALL TURNED IT INTO A HISTORY PIECE. UGH.

8:07 PM

there's a reason I don't follow these issues in great detail

Ok I need advice - my team lead is kind of "anti-planning" in that I'm expected to just do the minimum necessary to do something - in the minimum methods, classes etc, and we can come back and fix it later

but that just feels wrong

I want to break it into logical objects and keep it neat.

I also don't want to come across as being stubborn

user41796

How long do you realistically plan on being there?

is he deliberately choosing to sacrifice short-term code quality for some specific reason, and/or planning to add refactoring tickets to the backlog later?

I guess it's really obvious which of us has had multiple jobs

@Ampt now I'm listening to Reagan's "tear down this wall" speech. thanks. :P

user41796

@Ixrec I have worked in the worst types of that culture - and the answer is that you do the best you can to survive as long as you can all the while working on getting the hellz out.

8:14 PM

@Ampt tough shit. You see your choices; pick. Stubborn or team player. Welcome to industry.

@JimmyHoffa I don't think she would be too upset if I pointed it out

she's done a few similar classes and expects me to use those as a template

user41796

The caveat being to make sure they don't burn people for doing exactly as told and then later on blaming them for failing to see X, Y, or Z.

and If I do that, then there's not room for breaking things down like that

@GlenH7 project ships in march.

@Ampt you mean planned to ship in march ;)

this all boils down to what you think your current company/managers are like, how much they care about actual quality versus hacking stuff up and finding solutions vs blaming people, etc, etc

8:15 PM

well, yeah, you get the idea

user41796

@enderland No, I do think he meant that it will ship in March

I have a rough idea how much suggesting I can get away with at my workplace

@GlenH7 SHIP IT NAO!!!!!

@Ixrec I'm on week 3 or so of this project - so trying not to rock the boat too too much

do you have 1-on-1s there?

user41796

@Ampt If she seems reasonable, try a few tentative questions and see what the feedback is. If it's truly a hard and fast deadline, then that has to be met. But if there's some wiggle room in order to maintain code quality, then you'll see that in her response.

8:17 PM

@Ampt if you follow the instructions, will you be given shit later if/when it doesn't extend well? If you don't need to CYA, don't press too hard. If you won't come back to it and have to extend/maintain it later then YAGNI (If you're not absolutely 100% certain, then you will have to maintain it, because we all have to maintain everything all the fucking time).

That's it too - once I'm done with this, the likelyhood of coming back to fix it is low, but I also want to write good code wherever possible

If you'll have to come back to it later to extend/maintain it, request to implement it in an evolvable maintainable fashion. If your feet will be held to the fire later for how it evolves, politic to implement it in an evolvable maintainable fashion.

is the next maintainer likely to be a violent psychopath?

@Ampt Good habits are good, practice them with a comfortable frequency but not at the expense of political points (go along to get along will earn you points to spend later)

If you're only doing things shittily you'll build bad habits. If you're usually allowed to do things well, doing things poorly occasionally won't ruin your habits.

She also pointed out quite a few bad coding practices in the old project (this is a redesign) and said she didn't want to repeat those

so I think this is a chance for us to make sure that doesnt happen

user41796

8:21 PM

Pick a simple example that highlights your concerns and show it to her. Gauge the reaction and adjust accordingly

@Ampt sell it. Always practice selling your ideas regardless; it's the only way to learn how to be well received, which is absolutely necessary to get anything done according to your choosing in this industry.

user41796

If she blows up, then let it roll off you and don't worry about it. If she's receptive to that change, have another and slightly bigger example ready

Alternatively, knock back a glass of Bushmills and just start yelling at her about your grand ideas and draw diagrams on the whiteboard until she submits to your demands. They all given in eventually.

@JimmyHoffa I like this idea because Bushmills. BRB running to convenience store downstairs.

8:44 PM

-5

Q: Eelectronic parts catalog ecommerce

Im looking for an solution like this for our webshop http://utzon.se/upd/38/189/008 How is it built?

web-development api mysql websites

how do these people get jobs

user55340

@enderland elance: "I'll build it for you for $30" ... SE/SO: "I am looking for an solution like this. How is it built?"

@MichaelT but really, how do people function who can't use Google to at least start their search. I get some people might have actually had to learn things prior to Google but... now the basic threshold is so low yet still missed by so many

user55340

@enderland they are the lowest bidder. The question is "can they get someone to do it for them for free?" - and enough do that it is a profitable model for at least some.

Given a Match() method that returns all matches of a regex within a string, what would be the regex for an exact match across the whole string? What would be the regex for an "or" match, as in "this" or "that"?

user55340

@RobertHarvey /^(?:this|that)$/

8:48 PM

OK, so the ^ is StartsWith, and the $ is EndsWith, right?

user55340

Yep.

What does the ?: mean? "One of"?

user55340

I like putting them in there if it is a complete line match just so you know it is going to match the entire line.

user55340

(?...) is a non-grouping match.

And the colon?

8:49 PM

I should learn regex sometime, or at least, enough of the basics

user55340

if it was /^(this|that)$/ - "this" or "that' would be in group(1) if it matched.

Ah, better.

I think.

user55340

/^(?:this|that)$/ doesn't store anything in group(1). Again its a "I don't care about this, don't make the person reading this think I do."

No, but I could force the underlying code to use group(1).

user55340

You could also write it as /^th(?:is|at)$/

8:51 PM

Are the slashes significant, or do they just mark the start and end of the regex expression?

user55340

And if you had /^(th(?:is|at))$/ - again, this or that would be in group one.

user55340

I'm a perl person - they're part of my lexicon.

OK.

user55340

In Java, it would be Pattern thisOrThat = Pattern.compile("^(?:this|that)$");

And if you didn't want anything special, you could simply pass "this" as the regex?

If you just wanted (more or less) this == string?

user55340

8:53 PM

You could. Though, why do such if you were just going to do str.equals("this")

Because I don't want to have to throw a switch for the regex.

user55340

Building the Pattern (regex pattern, not GOF pattern) is actually compiling it. Really. Complaining the string into a state machine.

yes, regexes match literally, unless it contains regex operators, in which case you'd have to escape it first.

Escape it?

user55340

If you had some other funny things in there.

user55340

8:54 PM

Like if you wanted to match a markdown link...

Oh, OK.

I won't have anything like that.

user55340

/\[[^]]+\]\([^)]+\)/

user55340

Since [] is a regex special character, I need to escape it to match it literally.

user55340

Same with the () being special characters.

Yeah, I don't want to ever see anything like that in this particular scenario. That's the danger with choosing regex as a domain-specific language for this; the ones who really know it will get too clever for those who don't.

user55340

8:57 PM

That one isn't that bad except for the fact that its all straight lines and the occasional ^ and +.

@RobertHarvey If I want to match the literal string "(foo)+bar", I actually have to use the pattern /\(foo\)\+bar/. See Regex.Escape() in C# – special characters are \, *, +, ?, |, {, [, (,), ^, $,., #, and white space

once I read the full list of everything that all flavors of regular expressions are capable of doing

user55340

@Ixrec PCRE are much more powerful than classic RE. There's a neat paper on building the model for computation for that.

it's a bit ridiculous

was PCRE the flavor that had recursion?

user55340

cial.csie.ncku.edu.tw/st2010/pdf/…

user55340

9:01 PM

@Ixrec depends on how you define recursion.

there's a regex flavor that has an explicit recursion operator, \R or something I think

@Ixrec yes, and it's glorious – you can parse all LR(*) grammars with Perl or PCRE regexes

@amon In this specific application, the only things that we are likely to see are alphanumeric characters, and maybe dash and comma.

user55340

perl -lne '(1x$_) =~ /^1?$|^(11+?)\1+$/ || print "$_ is prime"'

most readable thing ever

user55340

9:02 PM

convert a number to unary. And then try to divide evenly by everything.

user55340

Incidentally, I used to play on a mud (Viking) with Abigail. Very interesting coder.

user55340

perlmonks.org/?node_id=50291

@RobertHarvey imo, the really useful bits to know (that don't nuke readability) are: capturing (...) vs non-capturing (?...) groups, backreferences (\1, \2) for find replace, character sets [a-zA-Z0-9], negated character sets [^a-zA-Z0-9], the start/end of line markers ^ and $, and of course the quantifiers: . + * {2,3}

everything beyond that gives me the "I'd rather write real code" feeling

@RobertHarvey If it's user-defined input, just put it through escape. If you're really sure it's that restricted, you can use such strings literally. But there's not much practical difference between Regex("^" + string + "$") and Regex("^" + Escape(string) + "$"), except that the latter is less fragile.

user55340

> Q: How do I get the date/time for this time tomorrow?
A: sleep 86400; print scalar localtime;
Did you see Ilya Zakharevich's response to that? Here it is:

This implementation is very pessimal speed-wise. Here is an important optimization:
sub tomorrow_date {
sleep 86_400;
return localtime();
sleep -86_400;
}

9:05 PM

rofl

Lightness Races in Orbit

o.O

"What's wrong with ^ that?" might be a good fizzbuzz replacement

Lightness Races in Orbit

technically, both solutions are wrong

I think that's the point

Lightness Races in Orbit

sleep x sleeps for at least x, not exactly x.

so if you want "this time tomorrow" that's a wrong solution

9:07 PM

@amon Did I say LR(*)? It's actually just LL(*), which happens to be the same grammar class as ANTLR, but with exponential complexity rather than linear run time.

it's also not clear what happens if DST transitions occur, or if the machine is on an airplane while the program is running

user55340

Abigail is the master of the what have you tried non-helpful answer.

Lightness Races in Orbit

right

ugh, EL&U is still such a cesspool

and people say SO is elitist

it's like someone took all the negative traits seen in SO contributors -- the childish, logicless "fights", the posturing -- and transcribed it onto people who are generally older, and who have generally performed more research into academic topics over their lives. They use longer words and they really, really want you to believe that they're much cleverer and better than you.

..... even when they're engaging in the typical sort of puerile internet behaviour you see from people like Vlad and Alf on SO

it would be amusing were it not so sad....

also, the answers attracted by the "what is a phrase/word for ..." questions are usually laughably wrong

user55340

The only thing they have to show off with is the language itself.

user55340

And thats why I also fight the naming questions so hard here - I don't want that to be an acceptable question here because it attracts the wrong sort of attention and doesn't produce people who are committed to being part of the answering community here.

user55340

9:20 PM

btw, @LightnessRacesinOrbit did you see the fun in that reclosed question?

user41796

@MichaelT Said more simply, they're boring questions for the experts we wish to retain in the community.

user41796

And boring for new / future visitors who will form the next revision of the community.

user55340

@GlenH7 very. Its the type of thing that hits hot questions, people come by, drop a turd in it, and continue on... and then people up vote everything and whee! Unicorn poo everywhere!

user55340

@GlenH7 which also never get searched and closed as a duplicate - they are the incarnation of "this question helps only one person."

user55340

Unless you're harvesting them for a word of the day calendar.

user41796

9:22 PM

ding! Fully agreed

more importantly, they're pure opinion polls

@MichaelT reclosed question?

user55340

@Ixrec This mess: programmers.stackexchange.com/questions/303249/…

user55340

@Ixrec Again, the Atwoodian model has failed there.

that's more the migration system being broken than anything to do with Atwood v Spolsky

Lightness Races in Orbit

@MichaelT nah?

9:24 PM

the core question of "is exec different from _exec" is still an SO question

Lightness Races in Orbit

I can't see deleted questions :(

user55340

@Ixrec different conversation thread.

user55340

> Haha! I can't stop laughing. Let me explain. I first posted this question on Programmers.se, which was then migrated on StackOverflow. That post was then marked duplicate of which @enderland is specifying. I decided to delete my question in favor of that one. So I merged my contents of my question to that question and delete mine. Check this revision to see how that question looked before my edit. I don't know how this question on Programmers.se reappeared. The question you people are saying the same is actually made same by me. – Santosh Kumar 5 hours ago

Lightness Races in Orbit

oh, that twat who copy/pasted an unrelated question without attribution

user55340

@LightnessRacesinOrbit yep. That mess.

9:25 PM

I think I voted to migrate on that one...

@LightnessRacesinOrbit I think he edited the dup to clarify the problem, which was his problem, then deleted his question so it looked like he copy/pasted

lesson learned: NEVER vote to migrate even if it's good enough for SO, it'll just be a duplicate anyway

Lightness Races in Orbit

@enderland maybe; not sure of the timings

user55340

Its a mess no matter what way you look at it.

Lightness Races in Orbit

I was pretty sure that wording wasn't his, though

it was (a) nothing like the first half of his question, and (b) actually intelligible.

9:26 PM

either way, a self-delete on a migration really shouldn't reject a migration

user55340

@enderland Time for an MSE post!

user41796

@Ixrec I almost always vote as unclear instead of migrating to SO. It has to be a really, really good question for me to vote to migrate

@MichaelT I hate posting suggestions on MSE

so do I, that one seemed to be abnormally well-specified

user55340

@enderland I'm working on it... I want the imaginary unicorn poo points.

9:27 PM

@enderland I kinda gave up on MSE after it became obvious there's no interest in fixing migration, which is the only thing I really want fixed

Lightness Races in Orbit

open.spotify.com/track/2LDiP3fyl2GmMeiVjmofQT

@MichaelT I dislike the whole system. it's basically, "someone posted this before and no one looks at it so de-facto rejected"

the core issue is that people on site X are deciding what questions go on site Y

user55340

@Ixrec This isn't so much 'fixing migration' which is a problem, but rather "this is a bug"

Lightness Races in Orbit

@enderland Isn't that the justification you lot give for not talking about renaming Programmers.SE?

9:29 PM

@MichaelT the delete un-migrating it you mean?

user55340

@Ixrec yep. That's how I'm going to pose it.

@LightnessRacesinOrbit SE holds the cards there, though, it's not worth my time getting worked up over it

@LightnessRacesinOrbit it's also the truth

Lightness Races in Orbit

@Ixrec Right, but in this case suddenly enderland doesn't like it ;P

you can care about many things. it's far more beneficial to care about things you can influence

9:29 PM

^

hence, we talk about close votes

speaking of things we can't influence

have we heard anything new about what Jon's vision/plan is for PSE?

Lightness Races in Orbit

> is this possible to bill a customer manually (through cron job in PHP) according to it's subscription plan (monthly/yearly)

lol

@LightnessRacesinOrbit nah, I'm very ok with it. SE's goal, as a company, is to make money - a lot of the needs of the company are different than the needs of the community as far as this goal is concerned

wow, I just saw a Skeptics.SE question with the comment "I think this should be moved to programmers.SE"

user55340

9:47 PM

0

Q: Self delete on target side shouldn't reject a migration

Aside from the giant mess of revisions, closes, migrations, reopen, reclose and delete that this question is (sorry, 10k Programmers.SE link - though other parts may be visible in the self deleted rejected post, and a wholesale rewrite of another question)... self deleting a post on another site ...

bug migrated-questions migration-rejection self-delete

user55340

The conclusion of A Group...

user55340

> Writing social software is hard. And, as I said, the act of writing social software is more like the work of an economist or a political scientist. And the act of hosting social software, the relationship of someone who hosts it is more like a relationship of landlords to tenants than owners to boxes in a warehouse.

The people using your software, even if you own it and pay for it, have rights and will behave as if they have rights. And if you abrogate those rights, you'll hear about it very quickly.

user55340

At times, I think that SE needs to listen to their board members a bit more.

what is this from?

user55340

@Ixrec shirky.com/writings/herecomeseverybody/group_enemy.html

9:54 PM

oh THAT post

come to think of it, I never actually got around to reading it

user55340

Give crunchbase.com/organization/stack-exchange/advisors a glance too... note the "More board members and advisors" - fourth one down.

« first day (1907 days earlier) ← previous day next day → last day (3087 days later) »

Transcript for

Nov '1523

The Whiteboard

General Discussion for softwareengineering.stackexchange.com P...

coffee-day horror-stories monads my-code-is-compiling scotch the-other-monitor

join this room
about this room

00:00

06:00

12:00

18:00

all times are UTC

site design / logo © 2024 Stack Exchange Inc; legal

mobile