« first day (1806 days earlier)      last day (2076 days later) » 
00:00 - 20:0020:00 - 00:00

12:02 AM
@JohanLarsson I don't think I'll have a problem. Its an LRU cache wrapper with the underlying implementation being a LinkedHashMap. But its one of those "you know, I'd just like to make sure I'm doing everything right" things.
screen -ls can get you a list of the sessions. Why not try parsing that list or parsing the output of ps? Once you have the sessions, it seems like a straightforward for i in If you make an attempt at writing the script and come back with problems, you are more likely to get help. From a previous meta SO post: "The first thing to understand is that Stack Overflow, and especially Programmers.SE are not code writing services. You don't just throw some a question in there with something of the form "give me the codez" into the text area and hope that someone will answer the question" — ssnobody 50 secs ago
@Duga nice false positive there.
2:20 AM
@WorldEngineer heh... you know you want to do this now:
@RobertHarvey I believe it was put on hold because expectations were mismanaged by my boss which led to customer frustration. — BAR 6 mins ago
Same as his question.
@MichaelT Not a false positive. Question is closable.
Duga identifies closable questions with like 95% accuracy.
@RobertHarvey what I meant was "that was not someone suggesting a migration and actually saying nope don't go there at all for this question"
Depends on if the 'positive' is the 'this post needs action' or 'this post is someone suggesting a migration to P.SE that needs action' - I was using the second positive in which case it was false. You're working from the first, which isn't a false positive.
2:31 AM
Just poking you in the ribs.
And today in "crazy British people": indiegogo.com/projects/realise-minas-tirith#/story
> We are an ambitious team of Tolkien fans who are passionate about creating a beautiful, inspirational and fully-functioning replica of Peter Jackson's depiction of Minas Tirith, as seen in his Lord of the Rings films.
> £1,850,000,000 GBP goal
(I do like how they're using the short scale rather than the long scale... its 1.85B rather than 1850 thousand million)
The long and short scales are two of several large-number naming systems for integer powers of ten, that use the same words with different meanings: Long scale: Every new term greater than million is one million times larger than the previous term. Thus, billion means a million millions (1012), trillion means a million billions (1018), and so on. Short scale: Every new term greater than million is one thousand times larger than the previous term. Thus, billion means a thousand millions (109), trillion means a thousand billions (1012), and so on. For integers less than a thousand million (< 109...
5:43 AM
@MichaelT yeah, just a lame joke by me.
@Jimmy would you call immutability & purity patterns? I would.
2 hours later…
7:59 AM
You are absolutely right and because regular expressions are mostly used by programmers, I think StackOverflow is the right place to ask this question. Hopefully, the community will contribute to my post. — nowox just now
8:45 AM
This question would be better asked on Programmers. StackOverflow is for more concrete, code-related questions. — sashoalm 16 secs ago
2 hours later…
11:13 AM
I'm voting to close this question as off-topic because it belongs to programmers.stackexchange.com — Anony-Mousse 12 secs ago
11:56 AM
So many companies are listing "open office floor plan" as a benefit. I really like collaborative work spaces, but I need at least a cube, if not an office, to be highly productive.
I really should start blogging about these things.
I've got a lot of things on a list, but I wanted a bigger list of things to write about before I started writing so I could always have a relevant queue.
12:40 PM
Happy Coffee Day
@ThomasOwens yeah, it's thoroughly baffling. Either they're all so clueless that they genuinely think it's a benefit, or it's just all marketing nonsense trying to sell it as a benefit so they can have all the management benefits without people complaining
@JimmyHoffa I do like the idea of collaborative spaces. I've suggested building some rooms that are smaller than a conference room, but have a whiteboard and a TV as a monitor so you can either use a computer or a laptop and work collaboratively in small groups (2-4 people).
1:04 PM
@ThomasOwens there was one of these for probably every 10-20 cubes at Employer^^^ - called them labs, they couldn't be scheduled on calendars, they were awesome.
@JimmyHoffa I'm not sure if they should be schedulable or not. I'd be disappointed to find them all used.
Size of an office, had a projector, a whiteboard, and a computer (though everyone would just use one of their own laptops), they were great for grabbing someone to go over something you were wrestling with or help someone or knowledge transfer between two devs doing a handoff etc
@ThomasOwens if they could be scheduled that would happen; there were enough of them they weren't all used ever
@JimmyHoffa Space is a little tight here, just due to the building and the fixed walls.
problem with scheduling is people fill the time to meet what they scheduled, plus scheduling means managerial types will grab them. Managers were never in them wasting time because what the hell would they do in one? Imagine producing something useful? Daydream?
Some of our rooms are only schedulable by an admin. Maybe that would work. If a team needed it for a day for some reason, they could get one and have it guaranteed all day. I suspect management wouldn't because it would be too small for a meeting that couldn't normally fit into an office anyway.
1:13 PM
@ThomasOwens yeah, it was the only place I worked that had it because it was a huge engineering HQ for the company; like 500-600 people in the building including pretty much every engineering department of various sorts.
Not a lot of companies have the space to do it, but the idea is sound and very helpful for engineering teams, I can vouch for that.
I should poke people about it.
A: Open letter to students with homework problems

DigitalRossI answer homework questions at times.1 I treat them like code-golf contests. Perspective #1: if I can do his homework for him all the way to graduation, then I have one CS grad that will be absolutely no competition. And he deserved that fate. Perspective #2: what exactly is my responsibility ...

1:36 PM
it happened just yesterday that moderators deleted three identical homework dumps, two of which were "inspired" by answers received by the first one (10K link) — gnat 5 hours ago
Happy Coffee Day Progs
Furiously mashes keurig button to make it brew faster
1:50 PM
@Ampt put a crank on the side to help it grind the coffee faster.
2:07 PM
(1) This site is not a forum (2) This is not the meta site (3) programmers.stackexchange.com may be suitable, but read the tour first — Sebastian Negraszus 34 secs ago
2:32 PM
@ThomasOwens How did refactoring your file processing to use a queue work out?
@RobertHarvey Lol I love that bobince is the one with the non jquery answer
Btw, @RobertHarvey did you see the stack overflow mentions in that oracle JVM talk?
@MichaelT link?
Two nights ago.
2:36 PM
Haven't seen it.
Search for "88,846" that isn't this message. Next link and on.
@durron597 Check out the "related", especially the last one.
It's amusing for so types. Quite interesting for Java coders who sometimes need to grab live wires.
@RobertHarvey beautiful
2:40 PM
@ratchetfreak also a video on AOT in Java.
And Jinq - yes that means what you think it might.
@durron597 Still profiling a few different implementations.
And dealing with some exceptions in a couple. I think it's just sloppy stuff on my part, though.
I have a few variations in mind that I intend to profile. One is the current implementation of reading each file and processing it. The other is to queue each file and then process the queue. Then I want to add different parallelization over each method. Threads per directory, number of threads to read from the queue, number of threads to populate the queue.
I have 2 weeks for the next software release to QA, so I'm going to take my time and measure and see what works. I may actually end up including different implementations in the package.
There is another StackExchange service for such questions — Szer 18 secs ago
@ThomasOwens cool
Exception in thread "main" java.lang.StackOverflowError :(
Need to debug my recursion, I think.
2:53 PM
@ThomasOwens Exception in thread "main" java.lang.StackOverflowError. Please contact your self referential support network.
Hm. I keep recursing into myself.
I probably did something stupid. I wrote this code last night before I went home.
transform to a loop?
Nope. I fixed it.
The loop would be convoluted.
I'm recursing down a directory tree.
add a heap of directories to be searched?
That would be another interesting approach...hm.
2:58 PM
then while((dir = heap.pop())!=null){...}
transforming from recursion to loop requires finding out what is stored on the call stack implicitly and make it explicit
@enderland your academic question is hot.
I could do it. I just don't think the resulting code would be as clear to a reader.
I suggest you flag it for migration to the workplace as this is a workplace issue. :-P
@MichaelT really?
Wow. Single threaded is slow...
But the recursion works now.
3:02 PM
Yes, it's hot. No, I'm not serious about the migrate flag.
(Can't see which reply a mention is for)
@ThomasOwens with a thread safe heap you just need to share a reference to the heap
@ratchetfreak Yeah. I usually do things in phases. First, single threaded implementations, then add threading on top of that once it's functionally right.
I find it easier to debug single-threaded things.
@MichaelT reply was for "is hot?" - go go upvote :P haha
A: What to do if i stumbled across a colleague in a scandal video online which seems to have been uploaded without her knowing?

JohnThis is Porn Police officer Sr. Lt. constable John Dontgetenough. Please let us know which site that video was on so we can take it down. Don't forget to mention the name of the video as well.

^^^ some job for ya
3:10 PM
@gnat It was 1 upvote.
I don't know who the hell upvotes crap like that
@enderland The people for whom it made laugh
@ThomasOwens "it's too addictive and too easy, and in the absence of any moderation, the community would do nothing but add and upvote the easy, fun stuff. This is why community moderators have real power; they need that power..." (The Trouble With Popularity) <--- @enderland this is expected
@durron597 there are now three answers there that are, ah, less than stellar
So throwing the files into a queue and processing the queue is faster. But if I only have one thread reading from the queue, then the number of files may be too much.
3:27 PM
here is the long version of my academic woes in case anyone is interested (@GlenH7 @durron597 ) - academia.stackexchange.com/revisions/51494/1
@enderland I agree, the only reason saving face matters here is to make it less likely that he will continue being unhelpful
If he is going to be unhelpful either way, just do what you need to do.
Yeah. i just would prefer to not have to take another 4 credits (probably $2k and a lot of time)
I think I'm going to escalate this to the university through formal processes anyways, part of me hopes he refuses to let me take him off my committee :P would make that process... trivial
3:57 PM
@enderland Is your thesis done?
@GlenH7 done is a very relative term, from my perspective? yes, but my perspective doesn't matter - only my advisors does
I understand that. Realistically speaking, if you gave a copy of it to the rest of your committee would they say "Yeah, you're ready to defend."
@GlenH7 ... I don't know - probably but I'm not sure
And maybe you've already considered this, but an option is to remove him completely, and have your new adviser put you up for defending your thesis
instead of futzing about with converting your thesis to independent study hours, just defend and be done with it
4:00 PM
that's generally pretty difficult in academia
Each university is different in that regards
in my case my advisor is the only university faculty with the expertise relating to my thesis work too
Ah, ok. That path isn't viable then
for all I know, I might be 100% of the way "done" with my thesis -- or 50% (unlikley to be this low but... I don't know)
Unless someone with more political clout was willing to take him on in a fight
4:02 PM
yeah. and realistically, I did the work/analysis like 2.5 years ago, I'm going to be rusty enough that it'd be hard to come across as super knowledgeable about the subject anyways
what did the ombudsman office and the director say to do for the next steps?
@GlenH7 I'm waiting to hear back from him, if I don't today I'm going to reping him and basically raise hell by saying "I talked with..."
I also emailed the other director for my 2nd degree program today
hoping to discuss, that might end this quickly
I want to frame the entire thing as one option: "change comittee and change program of study" not "change committee? and maybe change program of study?"
Yeah, getting it all done at once is wise in that regard
well and I want to make it such that if he refuses the program of study thing, he also refuses something that as I understand is flat out ridiculous to do - me changing committees
Probably wouldn't hurt be prepared to pay the 2k in additional class costs in case he's going to be a jerk about things
4:08 PM
yeah, I just need to know when - our semester starts a week from monday...
Ouch. That cuts it close for you then
well first week is pretty pointless normally
but yes
I am pretty sure I will file a formal complaint anyways
and if he refuses to change committees, that's a pretty dumb mistake on his part
Let's say I'm streaming messages through a series of objects (think Chain of Responsibility). Depending on the message type, I extract different attributes. The messages have overlapping attributes, sometimes, but the values are all the same. Do I have to worry about if multiple threads streaming messages to my classes? If the value is a single object or primitive type, I don't think so. Only if I'm modifying a collection.
@enderland Whee, ain't politics fun?
@ThomasOwens read only?
4:12 PM
@GlenH7 yeah. I'm just laying building blocks for things I guess should they (when they?) go awry
Assume they will. You'll be better off that when when they do. :-)
@GlenH7 The fields? They are writable. When a message of a given type is received, it pulls out all of the values that message type contains and sets the fields to that value.
So you could potentially have multiple threads writing to the same location?
@GlenH7 Potentially, yes. But keep in mind it's only a single object (String, BigInteger) or primitive type (int, double). I know I need to deal with concurrency if I'm modifying a colletion.
By same location, I mean calling an accept() method on the same object.
IMO you need to deal with concurrency any time you are modifying a shared object. Doesn't matter if it's a collection, primitive, or whatever
4:17 PM
Normally, yeah. But let's say I'm setting a field String x. It doesn't matter which message I pull it from, because String x is going to be the exact same value.
Otherwise what you're saying is Foo and Bar may each write to a single Baz, but that's okay because even though Foo and Bar pull the value they write from different places, that value is always the same across those different places
And that's an assumption that's very likely going to come back and bite you hard later on.
@GlenH7 Baz reads from Foo and Bar. That is, it has an accept(Foo) and an accept(Bar) method. If I read value x, y, and x from Foo and Bar, then x, y, and z will be the same in both. That's because I used values a, b, c, and d to determine that one particular instance of Baz should receive the Foo and Bar.
So Baz has an implicit assumption about the contract it provides with Baz.accept()
Because it's a chain of responsibility, I have Baz1, Baz2, and Baz3 as classes. There are multiple instances of each one, keyed to a value or a, b, and c.
Actually, that's bad naming. If I have A, B, and C as classes, I may have 3 instances of B created by A. A is responsible for looking at each message and handing it off to the right instance of B which may have 2 instances of C and will get the message to the right instnace.
4:22 PM
Seems like Foo and Bar should implement an interface (IBaz?) that guarantees Baz can pull from a single spot
@GlenH7 They do.
Cackles Yessss, yessss let the abstraction flow through you
But the messages flow through a processing chain based on values in the message.
@ThomasOwens And the area you're writing to affects future processing?
In reality, A holds a hashmap of B, which are candidates to receive messages and B holds a hashmap of C which are candidates to receive a message. Based on the value of b, A will choose the right instance of B to pass it to.
@GlenH7 Oh, no.
Those fields are immutable.
4:24 PM
I don't think there is a good place to ask this question. I think if you reword it to ask for a practice or advice on how to perform this review, then it would fit, but asking for tooling advice is off-limits for programmers and stackoverflow and code-review... — jessehouwing 18 secs ago
The IDs that I key off of to decide which child instance of the next step needs to handle the message is unique and immutable.
I don't think I'm clear on what part here you're concerned about with that flow.
There's the unique field that is used to determine what item handles each message. But there are other fields. I'm pretty sure I'm guaranteed that there won't be incomplete writes of data.
Is it something like this:
(msg) => A -> (b-msg) => B -> (c-msg) => C
message comes into A, it processes, creates a b-msg which is then sent to the appropriate B. And then B does the same thing for C
It's the same message that flows from A to B to C. Each one is extracting different content from the message. My concern is if there is two messages, msg1 and msg2 that arrive at the same instance of A (or B or C) at the same time that both affect field x of that object.
4:31 PM
@ThomasOwens I don't think that could happen. While you may have multiple instances of A, B, and C for processing, there will only be one instance of msg1
@GlenH7 That's true. But msg1 and msg2 are different objects.
Fora single instance of A, it can only process msg1 or msg2.
Back in a while - lunch
@GlenH7 Why?
I'm guessing it's because he's hungry
4:40 PM
Q: Can we have a guaranteed pipeline for responses from Stack Exchange?

durron597I think we should have some mechanism that automatically flags certain meta posts (both child metas and here) for Stack Exchange staff attention when they get beyond a certain point. The thresholds would be something like (subject to tweaking): Post is feature-request or bug Post has a score of...

@MichaelT @GlenH7 @gnat ^^^
5:01 PM
> Hi enderland,

Of course I remember you… but to be honest, I thought you finished a long time ago.
nothing like that response from my advisors boss (basically, the director of my other department)
@enderland lol wat
@ThomasOwens I am following up with the other graduate director now, who works closely with my adviser
@MichaelT Can I edit a link to the 6 to 8 weeks many memes of meta post?
@enderland Maybe you should have finished a long time ago?
@ThomasOwens yeah that's pretty much best case scenarior right now. lol
someone on my committee who is also the a director thinking "wtf you should be done already?"
5:18 PM
lol. I get a response back from my adviser about 30 min later, interesting..
@enderland posting on academia.se?
or talking to someone at your uni
@durron597 lol that'd be funny. but no, from my adviser
no i mean, 30 minutes after what
I got a response from my advisors boss 30 min after his boss (director) responded to my request
What is the state of the art in Rapid Application Development nowadays? ...
I have a use case where I need to create a GUI for several tables and generate a SQL script from it. In the old days I would have used Access for this, but I haven't used VBA in years, and this shop would be allergic to it.
5:22 PM
@enderland well, yeah. an advisor will ignore a grad student's email but not the director's email.
@RobertHarvey What do you mean by that?
@durron597 yeah. my advisers response is not to my liking, basically "you need to do more work to get the independent study credits"
Oh. I don't know what kind of bindings exist for that anymore.
@ThomasOwens What is the modern equivalent of Access?
@RobertHarvey I'm guessing .NET?
5:23 PM
Preferably in C#.
Yeah. I don't know where that stands these days.
Isn't Access still around?
MS SQL server express perhaps?
It is, but I'd rather not use VBA again.
SQL Server Express is a different tool.
SQL Server Express is just an RDBMS. Access is a development environment with a ghetto database engine and a ghetto programming language, but it's really good at quickly mocking up prototype applications.
ok I have no idea about databases, have no db in my life :)
did not stop me from taking a swing
If I asked Mason Wheeler, he'd say Delphi.
5:27 PM
I'd expect MVC + entity framework would whip something up pretty quick
but that was like 5 years ago when I had to do that.
and if you don't want a webapp or have that infrastructure it may be annoying
If I did that, I think it would have to go the Angular or Knockout route. It's basically a standalone application.
This is blatantly off-topic; it might be on-topic at Programmers. — Juhana 46 secs ago
@Juhana it's not at all on topic for programmers.SE. This question is not appropriate anywhere on the SE network in the current form. — enderland 21 secs ago
Ah, this is interesting.
Q: Hudson vs Jenkins in 2012

Vladimir BezugliyIn 2011 situation with Hudson and Jenkins was following(IMHO) - Hudson was a little bit stable, but development of Jenkins was a little bit faster. What is the situation with "Hudson vs Jenkins" now in 2012?

Same guy.
@RobertHarvey If you were programming in Java, I'd say spring boot...
Don't know the C# equiv
5:58 PM
Q: How to fight a kitty army?

AracthorLet's imagine an alien life form attacking Earth. Their technology and their number do not overtake very much human ones, but they have a very powerful weapon : when a human look right on one of them, he/she finds it so much cute that he/she looses every anger or will to fight, even if there is ...

I am sure @gnat will be along shortly to rant about the hot question list...
6:12 PM
Single threaded sequential file processing: 421.438 seconds. Single threaded queuing of files, then sequential file processing: 400.094 seconds. Multithreaded file processing (2*cores threads): 128.864 seconds. Multithreaded file processing (4*cores threads): 124.936 seconds. All of these reading files over the network. Still need to test n threads queuing files, m threads processing the queue. Any other possibilities I'm missing that may be more performant?
This is reading 1.5GB of data in 8781 files in 226 folders over a (pretty terribad) corporate network.
Your description of how they're implemented is pretty vague.
The first method is just recursive - go down a level if a directory, process a file. The second method is the first, but the files aren't processed until all the directories have been read and valid files queued. The last two spawn a Runnable for every directory and add it to the work queue.
None of those are what I suggested.
Which was your suggestion, again?
> Normal lambdas don't have beards.
6:25 PM
A: What is this smiley-with-beard expression: "<:]{%>"?

R. Martinho FernandesThat's an empty lambda using a digraph disguise. Normal lambdas don't have beards.

create a pool of worker threads that block on polling a LinkedBlockingQueue waiting for files.
Yeah. That's the one that I said I hadn't implemented yet.
Use another worker thread to recurse the subdirectories and offer to the blocking queue.
I think I am done with the Internet for today.
@Snowman diagraphs are stupid.
I am against any language feature that explicitly makes it easier to obfuscate code.
6:29 PM
@Snowman hot questions are probably less harmful at sites where average Stack Overflow lemming has no slightest idea of what to dump into answer box. When number of answers is under control, many things become more manageable
Digraphs were necessary in days of old. Backwards compatibility makes them a headache today.
@durron597 I'm about to implement your method and benchmark that against the other methods that I have.
I thought this was an EE question at first b
Q: Cat oscillating claws at me whilst being stroked

HaedrianI have a 2 year old cat which we got from a cat shelter and has been with us for almost two weeks now. He's very friendly and enjoys a good stroke and has never used his claws in anger. What frequently happens while he's being stroked is that he'll lie down and start kneading, but if the strokin...

@ThomasOwens k cool
6:34 PM
@durron597 The only difference is I'm debating making the number of directory reading threads configurable. So I may have 2 threads iterating through and finding files and 2 threads reading from the queue, or 1 and 3 or 4 and 2.
@gnat Thing about WB.SE is any nerd can throw out random theories as long as they are plausible in a sci-fi world.
@ThomasOwens reading directory contents is much, much faster than reading the files themselves.
@durron597 Even over a network and SAMBA?
In one use case, the files are local. In another, they are on a RAID array over fibre. In another, they are on a network share drive accessible over SAMBA.
@MichaelT when I saw "cat oscillating claws" and "stroke" I immediate thought of a rotor with claws, like some sort of medieval torture device.
The network share over SAMBA is going to be the slowest access out of any of them, I think.
6:36 PM
The pistons turn the crankshaft, turns the flywheel, which has claws attached that oscillate when a sufficient angular velocity is achieved...
@ThomasOwens Geez, can you do the processing on the other side and send processed json or xml back?
@MichaelT oddly enough, I know a cat that does this to me
or even protobuf?
@durron597 No.
@Snowman I know, I think I saw their rules. Key word is "plausible in a sci-fi world", it's not something that SO lemmings are capable of. I don't have 10K over there but it's quite likely that lame attempts at "fun answers" over there end up quickly deleted, like that "porn police officer" garbage over at TWP today, only making it easier for Community protection to kick in
6:38 PM
the solution was to have a blanket or some other extra layer over my normal clothes
@ThomasOwens lame
Mine will do that too.
I don't know what processing I'd do.
the same thing your app does
@MichaelT oscilating claws? I can't help but imagining someone attaching an osciloscope to their cat as it starts to vibrate, levitate, glow, and irradiate everything with varied spectrums of energy. They found Schroedinger's cat...
6:38 PM
just do that work on the remote machine
I'm just extracting data from binary files to show a summary of what's there.
Right; make the summary using CPU of remote machine and then just send the summary over.
Oh. There may not be a remote machine. It may just be a dumb network attached storage device.
World building requires strong writhing skills to get good rep.
@JimmyHoffa - I've had plenty of type errors before, but got my first kind error last night while working through Purescript By Example. Fun getting hit with errors previous languages couldn't conceive of.
6:39 PM
The only place I'm guaranteed any kind of significant processing power and application is on the machine running my application.
@MichaelT You think that MSE question will have any effect?
@durron597 four minutes after your question, two others got mod status tags.
@MichaelT and here I didn't even know that was a skill...
@MichaelT natural barrier to entry
@MichaelT lolz.
6:43 PM
Programmers has a bit of it... In that we downvote and delete poorly written answers, likely with greater frequency than other sites.
@MichaelT not workplace :P
@MichaelT yep, most technical subjects require some writing proficiency to communicate effectively, if only because it's so easy to create fatal ambiguity if you're not careful
@JimmyHoffa your famous monads answer I bountied a while ago suggests that you might actually have decent chance over there. And it was crazy enough, increasing your chances even further. "That's right, let's start talking about a monad..."
When is the movie version coming out? — Yannis ♦ Jul 30 '13 at 21:10
@gnat I've been writhing all my life, I think I bring a strong background to this aspect of World Building for sure
@JimmyHoffa your mission is to bring monads over there
6:48 PM
@MichaelT one of those was a twitter message earlier
@gnat sorry, I can only recreate over there with monads already brought...
@JimmyHoffa don't you want to save the world from imperative tyranny
make it pure
This question would probably be better suited for Programmers.SE. — Azar 13 secs ago
@Azar pros/cons are not good questions on programmers.SE... — enderland 39 secs ago
@MichaelT Jon replied. I'm not really satisfied though.
What do you use to track the time you spend on work projects? Do you use a personal time tracker? Which one?
6:59 PM
There should be a badge for 100 consecutive days in chat. It could the the first and only pewter badge.
@MetaFight 100? try 365! :P
(I didn't log into chat yesterday and felt the panic associated with fucking up the 100 consecutive days gold badge)
@gnat yup.
Well, I got a response from both bluefeet and jon same day
So I think "they don't care" is a little harsh
7:03 PM
Honestly, this sounds like "we're not going to give any visibility into which features are queued. We are only going to give feedback when either rejected or when they are completed." Is that your intent here? — enderland 40 secs ago
@durron597 looks like they took only second part of Atwood's Listen to Your Community, But Don't Let Them Tell You What to Do
Nobody tracks their time?
That's frustrating to me as a user because frankly SE has a hugely active userbase that runs the Q/A part of their business model (ignoring careers, which is the primary revenue part of SE). Not having a way for me to get feedback about which idea SE is planning on addressing discourages me from contributing ideas, since as is pretty known by most people, there are even in your example almost 250 very popular requests that have no official input from SE on (and that's just meta.SE, ignoring all site specific metas). — enderland 8 secs ago
@RobertHarvey we have a broad bucket based system
How do you deal with "this is taking too long"?
in what regard?
7:07 PM
@RobertHarvey what categories do you have for tracking?
What do you mean categories? At Employer^^ we could define whatever categories and "projects" we wanted, and had fine-grained tracking of time down to the minute. Here, I have one number in TFS that I modify on each task daily to indicate how many hours of effort I think I have left.
oooh, you mean more estimating than tracking
This question might be rather suited on the Programmers SE. — Kyll 14 secs ago
No, I mean tracking. Estimating is a whole nother problem.
"effort I think I have left" <-- what do you mean by this? it sounds like you are trying to estimate future work, not track actual work
7:15 PM
The whole idea of how you define products, programs, projects, and tasks and then coordinate and track them to completion is interesting to me.
@RobertHarvey my girlfriend is a lawyer, she has some software that lets her track billable hours down to the nearest 6 minutes, is that what you want?
@RobertHarvey I track my time.
@durron597 Yeah, that's essentially what Employer^^'s software does. I'm not going to buy his though, for several reasons. I just thought maybe someone did this with either a freely-available or low-cost tool.
@durron597 Does she use an app? I use one on my phone called Time Recorder Pro.
You can set up target daily/weekly hours, overtime rates, billable rates by project, rounding (I round to closest 6 minutes, per police). It can be as easy or as complex as you need. It has clients / projects / tasks...
It also syncs a backup to Google Drive on a regular basis.
@RobertHarvey used one at Employer^ - had one for Mac and one for Ubuntu.
7:20 PM
@RobertHarvey Mobile or desktop?
Desktop, I think
I think outlook has some time tracking feature.
@MichaelT It's not good.
At least, I haven't figured out how to use it.
That Time Recorder thing looks pretty good.
@RobertHarvey There's a free trial. The free one is almost fully featured. I only bought it because I wanted to support the guy,
7:21 PM
It's the "Journal" feature.
Outlook does have "Journal", but it's kind of lame. At least in 2010. Don't know if it's gotten better.
Journal entries have a duration.
Yeah. You can open a journal entry and track things. I may want to look into it more.
They have a start and stop timer too.
Looks like the Sexuality.SE closed beta fizzled out. Not enough activity, basically. Back to Yahoo Answers!
7:23 PM
Mmm... Not enough activity.
@Snowman They only gave it 16 days?
@ThomasOwens Look at the stats, though. Way under a healthy level of activity.
@RobertHarvey Hah, I just realized that.
You could say that Sexuality... "needs work" according to that Area51 page.
I would have given it at least 30 days.
7:26 PM
Questions started off pretty well but it severely tapered off near the end
I use standups for that. What did this person expect to do yesterday, did they do it?
what did they expect to do this sprint? did they do it?
Closing sooner than later is better. People have expectations if it stays open too long.
If you haven't had any in 30 days...
After about four or five days of many questions at the start, it would go multiple days with maybe a single new question
That is one where the linking of accounts might have hurt it.
7:28 PM
In outlook, can you batch remove categories from everything?
@MichaelT the solution is make another account and do not link it
too much work and barrier to entry.
True, too much for someone who wants to contribute just a little. And those are important people, because a healthy site will hook some of them into becoming regulars.
@ThomasOwens There are lots of free ones
7:32 PM
@durron597 Time Recorder has a 95% complete free version without limits as well.
@Snowman askapornstar.com gosh I hope that's not a real link
And it's the best mobile one I've found so far.
If only they had a desktop version...
@RobertHarvey I wonder if I could make a desktop version of it...
@durron597 Good thing I am on my home computer. That redirects... not to a Q&A site.
7:34 PM
I'd need to study it's import/export functions.
One of the new features is a master/slave capability for cross-device synchronization. I should probably email the author of the app first, but I could be able to write a desktop version.
I've been looking for a project idea.
@Snowman I didn't try it (obviously)
(I am at work)
@durron597 I figured you just typed it in. I don't care, like I said, I am at home on my home computer
@MichaelT: My answer is attempting to give you an idea of the scope of the problem. We could potentially open the Trello board to the public, I suppose. (Don't hold me to that; we'd have to agree to it as a team.) But I don't think that would get to the root of the problem because the real problem is that it's nontrivial to connect a feature request to the person who can address it. The solution is a much smaller codebase, a much smaller dev team, a much smaller user base, and a much smaller database of feature requests. We can do better, but we can't make our process completely transparent. — Jon Ericson ♦ 3 mins ago
A much smaller user base is the solution?
@durron597 Fewer users means fewer ideas get generated.
@ThomasOwens ... I know that.
7:39 PM
Alternatively, it could mean fewer user classes.
For example, MathJAX was added because of the math and science sites. If you eliminated mathematicians and scientists, you'd have no need for MathJAX.
You'd also have a lot less traffic.
7:56 PM
@durron597 I implemented your method. Right now, it has one producer thread (that reads directories and queues up files) and n consumer threads (currently benchmarking with 4).
@ThomasOwens Let's see if it finishes before i get my coffee :-P
135.634 seconds for my implementation with 4*cores threads of an executor service. 112.811 seconds for yours.
@ThomasOwens Your message appeared as I walked back in. Time?
Yours ran second, so I wonder if there was any JITting happening...
00:00 - 20:0020:00 - 00:00

« first day (1806 days earlier)      last day (2076 days later) »