« first day (1109 days earlier)      last day (3875 days later) » 
02:00 - 18:0018:00 - 23:00

user55340
2:27 AM
Workplace types, would this question be appropriate for a migration over there?
 
user55340
0
Q: How to apply to multiple jobs at the same (small company)?

TruthOf42There is a company that really appeals to me, with two open positions that interest me. Both are very much programming jobs, that I would enjoy. I think I might like one than another but am not sure. Is it bad, or seeming desperate to apply to both jobs? Should I apply to a specific one, but als...

 
user55340
3:26 AM
@JimmyHoffa you'll love this - in the 'language prettfier' from Google that SE uses - code.google.com/p/google-code-prettify/source/browse/trunk/src/…
 
7:41 AM
@MichaelT regarding dupes visualization we recently discussed, and topics may be worth it
 
8:14 AM
What's the correct way to handle a question a few days old that got (wrong) answers. Is there a legitimate way to 'reactivate' it? Is it allowed to just ask the same question again?
That answer was not correct, he didn't answer on my question, and i change mark on that topic... — Zuhan 6 mins ago
-1
Q: How to link in panel four pic to one. Only winForm

ZuhanI need to make 4 different images in one, and it will be in the panel. Panel size will vary from 180 to 320. I tried to do one main panel, and in her place 4, which are fixed by anchors... What I have I try to get panels on this style private void Form1_Load(object sender, EventArgs e) { ...

 
8:29 AM
@thorstenmüller re-asking isn't recommended...
5
Q: Ask question with wrong answer again?

DrewI asked a question on Stack Overflow, and got an answer, but when I tried the answer, it didn't help. Now the question is old and not getting any more views or answers, even after editing the original question to reflect that the answer wasn't helping. Should I post the question again, or let the...

 
@gnat ok, thank you
 
8:58 AM
@thorstenmüller by the way, cross-posting seems to be considered even worse than re-posting: meta.stackoverflow.com/tags/cross-posting/info
> Cross Posting refers to posting the same question over multiple Stack Exchange sites. Cross-posting is strictly frowned upon as it leads to fragmented answers splattered all over the network. If you spot a user cross-posting, please make use of Moderator Flags to inform the Moderators. Refer to cross-posting faq for more details.
for the question with the wrong answer, I think a reasonable approach for the asker would be to bump it by editing in details on how they tried mentioned answer and what went wrong
 
9:25 AM
4
Q: Why is this question good?

DariuszAfter a week of absence I went to last week hot questions. Found this one Java while loop - I'm lost with 37 upvotes. Howcome 37 people decided that that question shows research effort, is useful and clear?. I am being more and more surprised by Stack Overflow community voting. There are some...

9 mediocre answers stuffed into it in 20 minutes after posting the question likely made it artificially high on collider, attracting views and votes. It's neither the asker's nor voter's fault that it has gotten more attention than it probably deserves; it's just a typical effect of broken "hotness formula" — gnat 57 secs ago
 
9:42 AM
@thorstenmüller Adding a bounty asking for alternate opinions is a possibility if the poster of the original 'wrong' answer has no alternate views
 
10:04 AM
On an unrelated note, what do you guys think of the push for Flow Based Programming?
^ The two important links referring to what im on about
 
10:30 AM
wish one day similar attention would be given to bug in hotness formula which continuously spams across the SE network, promoting low quality content and permanently trolls over Programmers and Workplace, "making interesting and well presented problems look the same as non-constructive popularity contests..." — gnat 1 min ago
 
 
1 hour later…
11:55 AM
Hi! I'm a Ph.D. student and it's diffcult for me to reach "real" software practioniers for my research. However, I think that academic research should focus more on their actual work practices, which is currently rarely done. Our group prepared a short online questionnaire on how and why software practioniers (developers, testers, architects, etc.) use sketches and diagrams, which is available at st.uni-trier.de/survey. It just takes 5-10 minutes to answers the questions.
7
 
12:09 PM
Just answered your survey. I hope it helps.
 
Thanks a lot!
 
12:50 PM
@seba229 I responded. I also starred up your message so hopefully a few other people who come later in the day can see it and choose to respond.
 
 
1 hour later…
1:56 PM
just discovered two tags that look like synonyms to me, are they? with 27 questions and with 5 questions. Isn't their existence a violation of DRY principle?
2
 
@gnat "Why is this question good?" <-- What's the antonym of 'tautology' ?
 
@JimmyHoffa DRY?
 
@seba229 You may normally have trouble finding industry programmers, but not right now, that's what everyone here is more or less. You could probably also find a good amount of real industry programmers around various subreddits, may wish to drop your link there as well (I don't know reddit enough to point you to good subreddits for this)
@seba229 This questionnaire refers to 'Sketch/Diagram' but there's kind of two different types and I'm not sure what you're looking for. I sketch things constantly in my work on paper, whiteboards, and sometimes electronically, but that's 90% of the time for communication purposes talking about a concept with a colleague, only occasionally do I create formal diagrams to be used in documentation.
Is your questionnaire referring to formal diagrams for documentation, or off-hand diagramming just during a conversation?
> DRY is short for "Don’t Repeat Yourself". This paradigm advocates to avoid code and data redundancy.
Yep, dupes.
 
@BoltClock'saUnicorn keep believing that your precious SO isn't affected by spam coming from collidergnat 25 secs ago
 
text of the tag itself shows it isn't referring to some DRYSOFT DryWall2000 software or something, it's the principle.
@gnat Bookmarks are best at illustrating a chat log section I find
link to a certain space in time makes it trickier to gain context because there's a lot of stuff around it that's less relevant than what a bookmark can identify
@MichaelT You are aware of the source of the term 'burninate' right?
 
user20683
2:13 PM
@JimmyHoffa The antonym of "tautology" is "contradiction"
 
user55340
@JimmyHoffa Yep. Strong Bad.
 
@MichaelT You're referring to it as "apparently" a technical term made me wonder I mean, if you're familiar with trogdor you know there's nothing "apparently" about it, it is a technical term. Those villagers got technically burninated not apparently burninated.
 
0
Q: Tag merge / synonym request: [dry] → [dry-principle]

gnatPlease merge tag dry (27 questions) into dry-principle (5), and make them synonyms. Preference for master tag is because I consider dry-principle a more descriptive name. Tag dry wiki excerpt is DRY is short for "Don’t Repeat Yourself". This paradigm advocates to avoid code and data redun...

@JimmyHoffa yeah. Funny thing, iirc I first learned about how chat bookmarks work when you demoed it here
 
user55340
@JimmyHoffa Its more a in joke for myself... and if any of my co-workers from netapp happen to pass by. When I started out there, I was talking with a tech writer who was translating my jargon for the manager. At some point I said something like "and then we need to ui-ify that page" and he said to the manager "thats a technical term."
 
user55340
She gave him an "I'm not an idiot" look (she was technical enough). For years afterwards we'd make up jargon on the spot and append it "thats a technical term."
 
2:24 PM
@MichaelT ah heh
@gnat There should really be a privilege and a review queue for tag merging so we don't have to ask mods for it all the time; I wouldn't call it a 3k privilege, but at some level you should be able to vote for it (4 or 5 votes to complete, put in queue to let the votes gather) regardless of your rep in the given tag
 
@JimmyHoffa your feature request is status-declined...
88
Q: Can we allow 7.5K users to suggest tag synonyms without a score of 5 in the tag?

M. TibbitsFeature Request: Could 7.5K users have the privilege of proposing synonyms without the 5 upvotes per tag requirement? (in other words unrestricted synonym suggestion) Background: I would like to be able to suggest tag synonym suggestions. I do not however, have the required 2.5K rep on Stack O...

although there's a widely shared understanding that something is broken here...
35
Q: Does the tag synonym suggestion system work?

Jon EricsonI've never seen a tag synonym created via the voting mechanism and I see evidence that the system isn't working to create synonyms without moderator intervention. Does anyone have any figures about the number of synonyms that are create by votes as opposed to the number created by moderator fiat...

 
2:39 PM
@gnat Starting my morning with some Kurt Elling if you don't know him, great modern jazz vocalist, does some creative things with the standards but pulls it off rather than it coming out like a disrespect as many modern folks attempts at being creative with them tends to
@JonEricson why was it never tried to simply create a review queue for them? I was thinking this morning about how tag synonyms always require mod intervention on P.SE (and on lower volume sites than SO altogether this is probably always the case), and my thought was make it a simple privilege just like the others: 7.5k or 10k or whatever required, 4-5 votes required to pass it, and a review queue to ensure it gathers those votes, the last part was obviously key when I thought about this as tag suggestions would quickly be overlooked. — Jimmy Hoffa 1 min ago
Alternatively you require X rep in the tag for it to show up in your review queue (regardless of site rep or along with X minimum site rep) but as you mention, visibility is a requirement and review queues were invented for precisely this reason. — Jimmy Hoffa 23 secs ago
I know it's long dead, but.. whatever... it still makes sense
 
2:56 PM
@JimmyHoffa having this as a review process makes a lot of sense, actually
 
@JimmyHoffa Kurt Elling has been highly recommended to me by some fellow music lovers 5 or 6 years ago. Back then, I gave him quite a lot of listening time. Agree that he's good, although somehow he didn't become my new favorite
 
so if you were designing a database layout/etc, what is a good tool to do this?
 
@ThomasOwens Thanks!
@JimmyHoffa Thanks for your remarks. Our questionnaire refers to both, off-hand drawing as well as formal diagrams. Just think of the last sketch/diagram you created, regardless whether it was formal or informal.
 
3:11 PM
@gnat Well then here's your musician you've never heard of before for the day.
@seba229 Ok. If you want to get a more accurate bit though you may formulate two separate questionnaires later for the two types of diagramming; In industry they are very different. I sketch stuff for myself just when I'm trying to think through things all the time, you can't really speak to "revisions" with it because it's a creative activity like coding, you sketch a little then scribble some parts out and sketch some more then throw it away and sketch something else and so on and so forth
When formally documenting something for a presentation/advisement or generic documentation however, the whole process, end-goal, everything about it is different.
Often times the formal diagrams programmers make for that type of purpose are quite useless to the programmer himself and of minimal value to other programmers. They are created to fulfill a required level of detailed specification, but frequently that detailed specification is more about answering questions and less about communicating concepts
 
@JimmyHoffa That's are exactly the things we want to find out.
I mean "that are"
 
"Those" are :)
 
Ok, sorry :-) You know, many people in academia think that UML is the standard for everything. I don't think that's true.
 
It blows me away how good so many people who are not native english language speakers are.
 
@enderland No joke, people speak english way better than english speakers speak other languages. By a looong shot.
 
user55340
3:19 PM
@enderland Yeppers.
 
user55340
Just picture @JimmyHoffa trying to speak Greek...
 
@MichaelT "Gyro"
 
user55340
(well, he does Haskell, so not too far away...)
 
(I'm assuming @seba229 you are German based on the link? but other than your "that's" just now I would literally have had no clue you were not a native english speaker)
 
@enderland Thanks. You're right, I'm German. :-)
 
3:21 PM
@seba229 UML falls into the formal type I was speaking about that rarely is useful to the programmers themselves, they're almost always created for presentational purposes to answer questions from non-technical types (I will isolate the sequence diagram as the single technically-useful UML diagram type I've seen)
 
@JimmyHoffa @seba229 you guys have waaaaay too similar of hash pics.
 
@seba229 I was in Dortmund for a summer. Definitely a cool place!
 
user55340
@enderland stray apostrophe's are not a give away though.
 
user55340
From Wikipedia...
 
user55340
> Apostrophes used in a non-standard manner to form noun plurals are known as greengrocers' apostrophes or grocers' apostrophes, often called (spelled) greengrocer's apostrophes and grocer's apostrophes. They are sometimes humorously called greengrocers apostrophe's, rogue apostrophes, or idiot's apostrophes (a literal translation of the German word Deppenapostroph, which criticises the misapplication of apostrophes in Denglisch).
 
3:24 PM
@enderland I have to admit that I've never been to Dortmund :-(
 
@seba229 it's ok, it's not much to really go to - I enjoyed it because it was definitely not a tourist city :)
 
@MichaelT "deppen" this must be the root for the word "dope"
 
@enderland yeah, I think the whole Ruhr region isn't very touristic...
 
@seba229 nah, Duesseldorf and Koeln are pretty big tourist areas - though the eastern part? yeah... not so much :D
 
user55340
@JimmyHoffa etymonline.com/index.php?term=dope claims its from the diminished mental capabilties from using dope - ""sauce, gravy, thick liquid," from Dutch doop. Extension to "drug" is 1889, from practice of smoking semi-liquid opium preparation"
 
@enderland A friend of mine if from this area, we would never say that Düsseldorf and Cologne really bolong to "them".
 
user55340
@enderland Ahh... while short and maybe idiomatic for another language, you'll even find me doing that. I start writing something, change my thought, start wring something else and improperly proofread the new sentence.
 
user55340
In comments, it leads to some awkward wording that I can't go back and fix once I realize it later.
 
user55340
> Its not so much the 'get code from SO or get algorithms from P.SE' thats the issue that is the trouble
 
@enderland There is a map in the Wikipedia article: en.wikipedia.org/wiki/Ruhr
 
3:34 PM
huh I didn't realize that, I'd always assumed the entire area was part of it
funny, I lived there for nearly three months and never even looked at a map of what the ruhrgebeit was
 
@enderland :-) I think must of the Germans also don't know where the borders of the Ruhrgebiet are...
 
yeah I guess it's not really the scenic part of Germany, that's for sure
At least compared to other areas
 
4:04 PM
@seba229 took your interview. You guys looking to create new modeling software or just doing a paper on diagram usage?
 
user55340
Oded got one close reason in 45 seconds. Whee mods!
 
user55340
-1
Q: Why is CMake so Popular?

sklnWhy is CMake so popular ? From my experience it has many pitfalls or perhaps I misunderstood the purpose and it's not truly a write once build for all simply a crud tool for managing the compilation for each compiler/system. It lacks even basic functionality to check what compiler is being used...

 
@MichaelT lol. awesome
 
user55340
There are times in the day when the regular close voters are out of votes and I've noticed some mods pick up the slack with the obvious "close this" questions.
 
user55340
sigh (and I'm out of close votes)
 
4:17 PM
@MichaelT I'd love to help, but alas, my lowly rep prohibits me from picking up the slack. Back to drinking more coffee and pondering the mysteries of the universe I suppose
 
user55340
@Ampt As long as you don't go about answering recommendation questions...
 
user55340
-2
Q: Prefered IDE For Web Development

RoryPicko92I suppose this is a bit of a personal preference question, but, I just can't seem to find an IDE for web development that I like! My favourite IDE for programming in AS3 flash development is FlashDevelop for many reasons, but mainly the useful shortcuts it has. shortcuts as simple as ctrl+shift+...

 
nothing like sitting in on meetings related to your dev projects led by people who can't write a single line of code.... woooo!
 
@enderland I'll pray for you brother.
On the other hand though, Job Security WOoooooooooooo
 
thanks. fortunately there is "mute" and StackOverflow!
 
4:24 PM
@MichaelT Whoops
 
@MichaelT I kinda think anyone complaining about cmake should go through the process to build open source libraries for a while, you start realizing the ones with cmake are SO much nicer to build
 
@MichaelT he deleted it himself. Save the close votes for later I suppose haha
 
@Ampt Pondering the mysteries of the universe? You have been learning haskell!
 
@enderland I'll admit, makefiles are still somewhat of a mystery to me. I get the general gist of it, but like regex, the whole syntax is pretty confusing.
 
@Ampt yeah no joke. I think 99% of them are "hey I found this make file which worked, and I modified it to work!"
maybe more than 99%
 
4:35 PM
@JimmyHoffa As much as I want to talk crap on haskell I was thinking that it could be useful for my algorithms class next quarter....
 
user55340
@Ampt The original Makefiles were rather straight forward... I still find them easier to think about than modern tools.
 
We have some for work that just work recursively looking for cpp and headers and puts them in a big list and makes them in the appropriate order. It's pretty foreign to me.
 
user55340
target: dependencies
        command
 
$(BUILD_DIR)/%.o : %.cpp | $(BUILD_DIR)
	mkdir -p $(dir $@)
	$(CXX) -c $(CXXFLAGS) -MMD $< -o $@
thats a few lines from the one that I just keep re-using lol
 
user55340
to make a .o file in the build dir, it depends on the coresponding .cpp file and the existence of the build directory. To make this, make the directory, invoke the CXX defined for partial complaiation with the flags defined and write it out to the name of the target.
 
4:37 PM
like I said, I get the general gist of it, but I'm not sure which flags do what and some of the other notations like $< and $@
if it breaks I'll figure out enough to make it work but making one from scratch would probably send me to drink more coffee and learn haskell
 
user55340
flags are an issue no matter what build system you use.
 
user55340
The $@ and $< I'm not 100% sure of (I liked wordy non-macroed make files, but they do violate ).
 
lol 404 on the link @StackExchange
 
user55340
I learned how to use make back in college. Partial compilation saved much time (the unix boxen where mostly old sparcs... even old then).
 
user55340
4:42 PM
Once you learn how to use make, I even use it today for little scripts where it needs to push files around, only if something else was updated.
 
@MichaelT I usually just let my IDE do all that fancy stuff
Say what you will but it works pretty well for me
 
user55340
@Ampt That works... until you do a build. Builds & deployment are sections of the software development lifecycle that I actually really enjoy working in.
 
user55340
Saying "build from eclipse and push to prod" is not a valid option.
 
why is that?
I build a hex from my IDE and then use that to program the devices
 
user55340
I've got a different class path than you do set up in my IDE. I'll build different code than you do.
 
4:44 PM
@MichaelT It saddens me how poor publishing mechanisms still are in visual studio and just in general
You would think by now such an imprtant thing would have sufficient facilities for quality build/publishes
 
@MichaelT ah, I suppose that would make a difference for applications on a computer. Since I'm re-write the whole flash on a targeted device I guess I'm lucky in that respect
 
user55340
@Ampt Ok then, I've got different code not committed to SCM...
 
user55340
you checkout the code, and try to build - we won't build the same thing.
 
SCM?
 
user55340
Source Code Management. Aka VCS.
 
4:46 PM
Source Code Manager?
ah ok. VCS
 
@Ampt Build/Release engineering/Configuration management is all about repeatability and traceability, making sure that you have a technique that yields the same assemblies every time, and can be traced to a retrievable set of source code that can be grabbed and built repeatably
 
but if the makefile gets auto-generated by the IDE doing the building..
 
@Ampt Makefiles should be checked into source control
 
hmmm let me look
Yep, its there
 
user55340
(btw, I'm not picking on you on this... I've seen a "director of technical architecture" not understand the value of a build server and his team did "build in eclipse, push to prod". Baffles the mind.
 
4:48 PM
The importance of good build/release engineer becomes really clear as soon as you have a severe bug in some software that's 6 months old. Being able to pull that code and built it precisely as it was built for release is a necessity to safely know when you fix the bug you aren't introducing anything new otherwise you would need to do a full regression just to release a bug fix
 
user55340
If your build system is good, you can even pull the version information from the binary if you don't know what version the bug is in.
 
So what am I doing wrong here?
not understanding the make file so that I can build the source the same way?
Not storing the makefile?
letting the IDE generate the makefile for me?
 
Or when you have testers OK a build not realizing that what they OK'd had some code snippets not checked in from a devs machine, or had a library pulled from that users machine in a specific version no one else has and the deployment ends up coming from an entirely different version of that 3rd party assembly which breaks in production
@Ampt Iduno if you're doing anything wrong; just speaking to what and why build engineering is important. Controlled conditions are what matter: That means building on a build machine, the with a known-configuration
 
well we have a build server that does all the release builds for us... Does that count?
 
Does the make build link your code with libraries on your machine?
 
4:51 PM
uh... depends on which system we're talking about
 
user55340
(gah, nothing on this build of linux has rcs ident keys in it)
 
user55340
   f.c:
       $Id: f.c,v 5.4 1993/11/09 17:40:15 eggert Exp $
   f.o:
       $Id: f.c,v 5.4 1993/11/09 17:40:15 eggert Exp $
 
@Ampt So long as those are used for testing as well as release yeah, that's the general idea. (Did someone say you were doing something wrong? I don't know anything about that; just throwing my two cents in on importance of build engineering..)
 
if I'm building a module firmware, I'll just make a hex targeted at the specific embedded chip. If I'm making an application for the main system, I'll build it on a VM on my machine and push it over to the system, if I'm making a new firmware for the main system, I'll use the make server
@JimmyHoffa I guess I was confused as to how we got down this rabbit hole
 
user55340
19 mins ago, by Ampt
@enderland I'll admit, makefiles are still somewhat of a mystery to me. I get the general gist of it, but like regex, the whole syntax is pretty confusing.
 
4:53 PM
9 mins ago, by MichaelT
@Ampt That works... until you do a build. Builds & deployment are sections of the software development lifecycle that I actually really enjoy working in.
I just didnt bridge that gap mentally
implying that what I'm doing is... wrong?
 
user55340
It can work well in a development environment. There is the distinct possibility that something will mess up when a dev-style build is used for production (lack of ability to find out where the build came from, what version it was built from, stray code in the build).
 
user55340
I do a build locally, but when I push to testing, I go to another machine and check out the code fresh from git, and then use that to make sure none of my local changes and tweaks made it into the production/test build.
 
@MichaelT Aye, I'm always mega careful with my checkins and when I'm going to do a merge/branch/dev or qa environment build, I look through everything and shelf any changes I want to continue working with, then undo everything and manually delete my entire local source repo to get completely fresh and build from exactly what's in source
 
@MichaelT there are people that don't do that?
 
user55340
The other thing is, right or wrong - its not something that one person can easily change. There are often processes around how things are built that have great amounts of inertia. For the software at previous employment, we basically had half a dozen people push down management's throats "we build on a build server"
 
user55340
4:59 PM
11 mins ago, by MichaelT
(btw, I'm not picking on you on this... I've seen a "director of technical architecture" not understand the value of a build server and his team did "build in eclipse, push to prod". Baffles the mind.
 
user55340
@Ampt yea...
 
@Ampt Tons.
 
user55340
He's even had bugs reappear because dev 1 didn't pull the code dev 2 checked in or dev 3 had local on his machine.
 
user41796
@MichaelT Check out the revision on it:
 
user41796
-2
Q: How to use CMake

sklnIs there any way to define compiler settings (ie for visual studio) without actually using visual studio's (bad) abbreviations for the settings ? How do you identify the compiler without doing a string comparison of a CMake macro ? How to split cmake files without a single one being 1000s of li...

 
user41796
5:01 PM
Night and day difference. How they decided to pull a question out of that rant, I don't know. For that matter, why didn't they lead with that question
 
user55340
@GlenH7 Much better, migrate to SO.
 
user41796
@MichaelT I'm still dealing with the WAT?! from the two completely different versions
 
The problem is to many non-technical people the whole of software development is black magic voodoo, and they can't fathom that the simple task of compiling is in any way trickier than creating the software itself, why would they need extra resources, it's all the same software after all regardless of who or where it's built. Right? And top that with many devs who feel like they can do it perfectly so they're not worried about the risks
 
user55340
@GlenH7 I'm glad he was able to learn how to ask an actual, answerable, question.
 
user55340
5:04 PM
@JimmyHoffa That software now, the manager wouldn't dream of doing a deployment from anywhere other than the build server.
 
> A decade of experience makes you a teenager
>
> After going through their 2-year-old phase, programmers eventually have to go through a stupid-teenager phase. All this month I've been hearing sad but unsurprising news stories about teenagers getting stuck on big rocks, being killed falling off cliffs, or dying of exposure. I'm actually lucky the same didn't happen to me when I was a teenager. It's just a bad time for us. Even though teenagers are old enough to understand the warnings, **they have this feeling of invincibility that gets them into trouble and often mortal peril.**
 
user55340
(the director of TA is still resisting though)
 
5:16 PM
@JimmyHoffa how do you feel about Scala as a functional language as opposed to something like haskell?
(you're my go-to FP expert now FYI)
To me it seems like a functional programming language that is inherently tied to java for better or worse which means it's kind of in that grey area between FP and (Regular programming?) whatever everything else is.
Imperative* is the name for it apparently. So it seems like it's somewhere between Functional and Imperative
 
@Ampt I generally suggest against using any functional language that allows standard imperative programming until after you've learned the heart of the functional style in something haskell or a non-imperative LISP
 
@Ampt Scala is billed as a multi-paradigm language.
 
@JimmyHoffa I'm not saying that I'm looking to use it, but I'm interested in anything that's more approachable. Just wanted to see if you had any knee-jerk objections to a language that's billed as both
 
There's tons of functional languages that also let you code imperatively, most are multi-paradigm where they're either more functional than imperative or more imperative than functional, the problem with them however is you won't learn anything new, you'll just happily and pleasantly go about coding just like if you were writing python
 
user55340
One of those 'pragmatic - do it however' that borrows heavily from fp, but is still a multi-paradigm language.
 
user55340
5:22 PM
If you want fp on a jvm, go for clojure.
 
@JimmyHoffa I remember reading about that somewhere, the guy was using some functional programming language with 0 of the functional parts and just couldn't figure out what was different about a 'functional' language
 
@Ampt The approachableness is the problem is my point, unfortunately it's the brain-bleeding mind-numbing pain of pure functional programming that teaches you to see something entirely new. As soon as you can simply write statements in order to be executed one after the other you will avoid the whole point
 
@MichaelT do you guys really get enough questions here to run out of close votes consistently?
 
Most languages are getting functional constructs. Java 8 is, C# has them, Python, Ruby, and Scala are all multi-paradigm.
 
user41796
@enderland yes. I routinely max out the review queue. And it's trivial to find additional ones during the day that need closing.
 
5:24 PM
@enderland they are very active with closing questions
 
7
A: From a high level programming perspective, where does the 'different-paradigm' barrier between C# and F# really kick in?

Telastyn I'm wondering where the 'different-paradigm' barrier really kicks in? There largely is no "barrier". You're free to write non-idiomatic F# all you want - just like you could program C# for ages and not really practice object oriented programming. The only case I can really see forcing your ...

 
user55340
@enderland People are dredging up old questions all the time - either close vote expeditions, or new person answers an old question.
 
user55340
You hit one "blah" question and then look at the related and see 5 more there that need a cv...
 
@JimmyHoffa I think that it may have been an f# question. He was using it, but in a typical imperative fashion not making use of any of the F# parts
 
ah, didn't think of the close vote queue...
 
user41796
5:25 PM
@Ampt I'm now #1 in the close review queue. @MichaelT is #4.
 
I am length -1 hahaha
 
user41796
programmers.stackexchange.com/review/close/stats - had hoped that would expand. Oh well.
 
user55340
As I see it now, the queue has 8 items.
 
user41796
@MichaelT 21 by my view. 101 if you log in anonymously.
 
yeah I see 100* over here
 
user41796
5:32 PM
@Ampt You should get cracking on those.... ;-)
 
Jeez, would you look at the time? I've got to go.... ermmm..... Learn Haskell. Yeah, that.
 
@Ampt I just linked you to it :P But it wasn't "not using F#" it was: Writing statement 1 followed by statement 2 to sequence your steps, that is the definition of "imperative" programming, until you stop doing that you won't learn the lesson of FP.
 
user41796
@Ampt at least closure would be involved
 
@JimmyHoffa I'm not sure that was the exact question. He was literally using only the C# notations. No monads, no nothing.
 
you could realistically learn FP in any language that allows higher order functions so long as you refuse to write any functions that executed more than a single statement
@Ampt Trust me it was. F# doesn't have C# notations
F# also barely has monads
 
5:34 PM
sounds like unit testing to me haha
 
@Ampt Nope, it's declarative
when you declare what something is, it doesn't take multiple steps, make sense?
sum [] acc = acc
sum arr acc = sum (tail arr) (head arr + acc)
 
None of it makes sense to me. I need @Yannis to interpret this greek you're speaking
 
user55340
@GlenH7 I still think this would be a good idea...
 
user55340
4
Q: Notify users of possible reviews on toolbar

MichaelTI am under the impression that increased participation in the review queue is, on the whole a good thing. It encourages people to feel an ownership in the community and partially responsible for the site as a whole. On Programmers.SE, I've seen what I believe to be an uptick in flagging to clos...

2
 
user41796
@MichaelT I like when edits pop up, so I think it's a good idea
 
5:39 PM
@Ampt All the same understand this: Don't look for something more approachable to learn FP, if you want something for practical use purposes sure grab something more approachable, but if you want something that will teach you FP make sure it's a non-imperative LISP or haskell (not sure of any other languages that disallow imperative coding)
erlang comes close-ish, prolog as well but then prolog is an entirely different thing. very interesting but not at all the same
 
I was just looking for your expert opinion on langauges that tried to do both as opposed to strictly functional ones
 
@Ampt They're great for use
I'm writing F# right now. Though I gotta admit, and I think @jozefg has echoed the same sentiment: Given the opportunity to code functionally in a language, I really don't find myself ever falling back to imperative in it.
multi-paradigm languages are a boon for imperative programmers, and a boondoggle to functional programmers, to me they feel... half-way right. But for production use there's benefit to the fact that you get a wider swath of developers able to pitch in vs. a real declarative FP language
 
So the attraction of multi-paradigm is that they're actually useful. Got it.
 
@Ampt The attraction is that more people know how to use them than FP ones
like I said, an FP programmer generally doesn't fall back to imperative in them, so to the FP developer they're not particularly useful
One of the problems with multi-paradigm languages is they tend to cripple bits of their FP facilities just because it wouldn't fit so well with their imperative ones, so you end up with things like F# requiring the rec keyword, and outright disallowing recursive types (which is infuriatingly stupid for a language that claims to be functional)
 
user55340
@JimmyHoffa a FP programmer dosen't mind the perl code @sorted = map { $_->[0] } sort { $a->[1] <=> $b->[1] } map { [ $_, -s $_] } @files;
 
5:48 PM
Scala suffers that sort of thing less than F# from what I've seen, the main pain in scala is the type inference sucks because of the inability to use algorithm W
 
user55340
Well, the theory behind it at least. (the -s is 'size of file on disk - an expensive operation).
 
user55340
But I'm sure they're less than happy with things like the = and loops and the like.
 
is $_ some sort of variable?
 
@MichaelT I don't see an assignment in there. makes perfect sense to me, you're manually doing a zipWith
 
user55340
@JimmyHoffa its the general 'a = b' that I was referring to. Mutables and the like.
 
5:52 PM
@MichaelT <=> is an assignment?
 
user55340
@Ampt In perl, $_ is the default variable. Within a map, its the current thing being worked on.
 
user55340
@JimmyHoffa The tie-fighter is a numeric comparison / equality. Similar to 'strcmp' in C or 'compareTo' in Java.
 
user55340
Excepts its a number, not a string (cmp is the string comparison / equality operator).
 
user55340
And I was more thing of '$foo = $bar' rather than '$a' and '$b' -- those are magical sort variables.
 
Tie fighter. Coolest operator ever. Brb, learning perl.
<=> pew pew. Die rebel scum.
 
user55340
5:55 PM
The spaceship operator, written , is a binary relational operator that originated in the Perl programming language. Other languages, such as Ruby and Groovy, also support the spaceship operator. Unlike traditional equality operators, which will return 1 (true) or 0 (false) depending on whether the arguments are equal or unequal, the spaceship operator will return 1, 0, or −1 depending on the value of the left argument relative to the right argument. If the left argument is greater than the right argument, the operator returns 1. If the left argument is less than the right argument, th...
 
Think wikipedia would be displeased if I changed that to Tie Fighter operator? I'm just trying to get people interested in computer science.
 
|-o-|
 
user55340
Then there's someone who did a super spaceship operator - perlmonks.org/?node_id=45627
 
or
<-o->
 
user55340
            1. < 0, push @ tnirp => <<=>> <= print @ hsup ,0 > .1
Just another Perl hacker                             rekcah lerP rehtona tsuJ

                              ; print @ tnirp ;
 
5:59 PM
    <=>
\  . ||.   /
 \        /
  \      /
   \    /
    ---
 
@Ampt Thanks. Currently we are writing a research paper on the use of sketches and diagrams, but we also have an idea for a tool.
 
@MichaelT I presumed it's behaviour was that, so not clear where you were referring to disliking the assignment in your snippet
 
02:00 - 18:0018:00 - 23:00

« first day (1109 days earlier)      last day (3875 days later) »