« first day (1621 days earlier)      last day (3666 days later) » 
00:00 - 22:0022:00 - 00:00

00:05
@JimmyHoffa lets shoot some baddies. you know you want to!
My linked-in profile, in case anyone is interested in connecting.
00:21
@psr yeah, I can see plainly how left recursion could be trickier, but I'm wondering if you can do backtracking with right recursion at all
It seems to maintain left associativity you have to define the grammer in an upside down graph with a start for each specific terminal rather than letting it recurse into the terminals because it'll never recurse left to maintain your associativity (this is my understanding of reading the literature on wikipedia on rewriting left-recursive grammers to right-recursive grammers while maintaining left-associativity)
With left recursion how you risk infinite loops is plain to see
user55340
00:57
For chem engineer types: things I won't work with tag on a blog. How about HOOH. Yep. Hydrogen peroxide. What if you made that HOOOH or HOOOOOH instead? Sounds like fun.
psr
psr
@JimmyHoffa You can set it up to match a list of expressions, which is passed to your handler function as an array. You can then put the array together left-associatively (that's how I did it, may not be optimal).
user15026
@MichaelT hahaha I have read most of those, they were delightfully horrifying
user55340
Oh no. Free house has peach apple cider. This does not bode well for my wallet.
user55340
01:31
@psr yeah so this is what I was thinking - you had to create a left-associative list of entries for the terminals at the entry point rather than with a left-recursive grammar you would only need a single entry point and let it fall/select to the left towards the terminals on the bottom of the graph. It's like making the graph upside down with separate roots for each terminal rather than one root
that's what it more or less describes in the wikipedia literature
(or how I parsed it)
 
2 hours later…
user55340
03:22
where snow clearing and computer science intersect. The snow blower problem (spoiler - it's NP complete)
04:06
Does <input type="datetime"/> work in Firefox 35? No, but it doesn't work in the latest version of Chrome either. Sure, avoid features that are not in the latest browsers. But seriously, is browser obsolescence really a problem anymore? If I want to, I can set Firefox up so that it updates to the latest version automatically, and I think this is the way it comes out of the box. Same with Chrome. — Robert Harvey 3 mins ago
 
5 hours later…
09:35
Hello, guys! I have a quick question which is probably answered in some FAQ, but it is not easy to find. Would anyone be so kind as to help?
I have an answer on "programmers" which was upvoted and accepted at some point yesterday, and today it is not accepted anymore, another answer is accepted. It is not that I mind, but I just want to know how the site works. So, is it possible to accept one answer and then change your mind and accept another?
question answered elsewhere, please ignore.
Indeed it is possible to do that.
 
4 hours later…
14:10
Is there a generic term for the idea of writing message formats in a way that is understandable by some kind of tool that can then generate the interface libraries you need to parse said messages into some kind of reasonable interface? Instead of a table in a document that defines the format of a message, it's a file that is both human readable/understandable, but also machine understandable to handle that message exchange?
I'm thinking along the lines of Google Protocol Buffers. A generic name for tools such as that.
Wikipedia calls ProtoBuf a "serialization" method, but that's not it. Serialization doesn't have the requirement of the ability to provide both a human-readable documentation and machine consumable method of producing a library.
14:24
@ThomasOwens So you're asking about Domain Specific Languages for specifying message schemata?
@amon I agree with "Domain Specific Language"
Domain specific language doesn't seem specific enough. Plus, it doesn't really imply the relationship between a data file and a tool to consume it to produce libraries. Maybe there isn't a specific term for this type of thing?
14:41
Perhaps “A schema language from which a tool can generate glue code”. The term “schema language” exists (e.g. XML schema languages such as RELAX NG or DTDs) and is apliccable here, but doesn't necessarily imply that code will be generated.
On a whim, I created an SO careers profile. The first box is "liked technologies and disliked technologies". I put and in the disliked box. The first two jobs that appeared in the StackOverflow sidebar after doing this are a C# job and a ruby on rails job. hmmmmmm
14:55
SE is calling to you
@enderland <ghost>Learn new languagessss..... ooooooooooooooooooooooo</ghost>
15:13
in Duga's Playground, 3 mins ago, by Duga
This site is more to do with programming questions, not really about how to teach compiler use. Perhaps hereryyker 22 secs ago
15:31
@durron597 The true nature of technology careers shows itself. You'll do it, and you'll like it, muahaha.
@RobertHarvey It'd be good to have more black walnut experience.
@durron597 cherry is a much prettier wood though than walnut
16:15
My wife was asking what I was thinking about during breakfast, so I tried to explain this using a "pegs and holes" metaphor.
"If I have a container with a round peg and a square peg, I can derive a new container by adding a star peg to it. But I need to put the star peg into the machine that makes the container, and I have to do that with another machine that looks at the containers and machines to find the hole with the right shape."
Ugh, the language specification is almost easier.
16:29
I don't understand exactly what he's looking for, but sounds similar to what I've used static generic classes for...
He's trying to figure out how to get a derived static method in a generic to fire, but T.SomeMethod won't compile and there's nothing to cast because it's a static method from a static class.
Normally, you would have derived object instances to call against.
So basically, you need typeof and a switch statement. Or you can call the static method with Reflection.
0
A: How to design similiar purpose static methods

Jimmy HoffaYou could create a static member for a generic static class that has the function for each subclass type. Just make each sub class have a static constructor that sets the appropriate action for how it does processing. public static class ProcessDt<T> { public Action<T> ProcessAllFromDt { get...

@RobertHarvey Why would you even try to do it that way, rather than just making an instance?
Does that work for what he wants you think?
On reflection (no pun intended), it seems to me that passing around object instances probably makes more sense than passing around types.
16:43
@RobertHarvey the thing is, for his specific behaviour the type is a predicate - he's talking about doing an if against the type for everything. Which is why I encourage using a generic because that causes the type system at compile type to encode the predicate for you
whenever you're using a type for a predicate, start thinking about how you can use the type systems features to dispatch the behaviour for you, it'll be more performant and easier to maintain than explicit conditionals
Yes, but you can't dispatch without an object.
You can't call T.SomeMethod()
@RobertHarvey can to, so long as a constraint guarantees it's there
It won't compile.
will to
Aw, c'mon. It's a generic type parameter. How is the compiler supposed to know about that at compile time?
16:46
public static void DoStuff<T>(this T target) where T : ClassThatHasTheMethod { target.TheMethod(); }
That's what generic constraints are for...
Yes, but he wants to call the method on a derived class.
they tell the compiler that the T will have these features, and that the compiler must blow up anytime something is handed to the method that does not have them
He already has the constraint, but it's on a base class.
@RobertHarvey ?? Yes and? the T : ClassThatHasTheMethod is a constraint that means inherits from, not is
so make the constraint on an interface with the method, or an abstract class with the method, all subclasses will have the method with sub class specific behaviour which will be called...
I think all those things imply object instances, but I could be wrong.
16:48
no
none of them do
if all he wants is to use generic constraints then my answer is terribly overcomplicated
hell I'm deleting that answer, I suspect he wants a simple abstract method and doesn't even know how to do that...
Don't delete it.
Look here first:
54
Q: Calling a static method on a generic type parameter

Remi Despres-SmythI was hoping to do something like this, but it appears to be illegal in C#: public Collection MethodThatFetchesSomething<T>() where T : SomeBaseClass { return T.StaticMethodOnSomeBaseClassThatReturnsCollection(); } I get a compile-time error: "'T' is a 'type parameter', which is not v...

why does he want to derive a static method ??? It makes no sense.
He doesn't understand static
You never ever have reason to do that - it logically doesn't make any sense.
Because he doesn't want to have to instantiate an object to call the method. He understands it.
@RobertHarvey no no, the point of derivation is to have an object that could be any of multiple types, if you know the type ahead of time, then there's no reason for this nonsense. In his case - he won't have an instance, so why the hell would he use derivation?
Why does he want those methods on those classes when those classes will never be there when he uses those methods?
He just wants static methods, tell him to put them all in a static class.
Really, I think he just wants to avoid the switch statement that would be required.
16:54
Either he will have an instance so he needs abstract methods, or he won't have an instance so he'll know the type. (truth be told he will have an instance, all his code examples have shown this to be true)
@RobertHarvey there will be no necessary switch statement...
He wants a factory that calls the same method regardless of the type. In other words, the behavior doesn't change between cases.
Well, not outside the called method, anyway.
He can still write different behavior in the derived methods, but the signatures are all the same.
@RobertHarvey there will be no derived methods without instances
public static myFactoryMethod(T object)
{
    switch(typeof(T))
   {
        case one : TypeOne.SameMethod(object);
        case two : TypeTwo.SameMethod(object);
        // Where TypeOne and TypeTwo both derive from the same abstract type having virtual SameMethod();
and the word factory means nothing
factory is used for constructing objects
he's clearly not doing any sort of construction
maybe my answer is right... fuck I have no idea what he's doing, he's totally vague and doesn't seem to understand some basic facts of the type system
never switch on a type, that's what generics are for
which is why I wrote my answer because he mentioned switching on a type
sounds like an X/Y problem, but if he just wants to avoid switching on a type my solution is correct
See my code above.
16:59
@RobertHarvey stop saying stuff about virtual static methods - it simply makes no sense
Well, it doesn't, really.
Note that in a functional language, this sort of thing would probably never come up.
@RobertHarvey no because we don't embed our functions in our types
Exactly.
It's an X Y problem anyway.
yeah
Perhaps my answer will make him realize he doesn't know what he wants himself
My coworkers are discussing whether or not functional programming makes sense for large scale implementations
it's like The Whiteboard -- IRL!
17:03
@enderland Did you read The Parable of the Two Programmers starred earlier?
@enderland and tell me, does a single one of them actually know a functional language thoroughly?
@RobertHarvey I don't recall
@JimmyHoffa Probably not? I'm not sure
enderland goes to read while IT "fixes" his computer"
Highly unlikely. Most decisions people make about FP ("It's simply not viable for large scale") are wholey uninformed
which is sad. It's mostly because people assume it's quite similar to what they already know so they figure their understanding is sufficient for a proper analysis (I figured as much before I really picked up Haskell)
@RobertHarvey That's rather interesting
@ryyker - this would be off-topic on Programmers as well. — GlenH7 41 secs ago
user114359
17:16
Does anyone know why sometimes when I type @ and someone's name I 1. don't get the suggestion window to pop up and 2. the whole @ thing does not end up in my comment?
user114359
So far today this happened to me with comments attached to two separate questions.
@Snowman you can only notice one person in a comment, trying to do it a second time or a second person in the same comment will just bail
user114359
@JimmyHoffa I know I am am talking about just one person
user114359
Like right there sometimes the @ JimmyHoffa would vanish
Try harder
user41796
17:19
@JimmyHoffa vodka and dinosaurs this early?
Raptor tequila
user41796
@RobertHarvey Oh, I'd pay good money for that...
user114359
I keep forgetting to ask (and I keep forgetting to sit in Whiteboard when I should be working too) but @JimmyHoffa can you tell us where you are buried?
@RobertHarvey Crustacean Vintage?
17:21
@Snowman Wherever it is, continuation-passing style was involved.
user41796
@Snowman you mean here in chat? Or in comments on main?
You also only need a few characters @sno
user55340
Wine and Dino. Pearadactyl anyone?
user114359
@GlenH7 it normally works fine here but normally in comments it fails
user41796
Two different systems
17:22
@MichaelT Dino and Dash - when you throw a chicken into a restaurant and run away?
user41796
If someone has edited a post, that will sometimes show up in the prompt. The author of a post will also show up, but the system removes it before submitting
user41796
And then you can get prompts on anyone else who has participated in that particular comment thread
user55340
@JimmyHoffa preferable over Bore and Boar.
@MichaelT trephaning a wild hog?
user55340
This meeting is boring. Hey look! A wild pig!
user114359
17:26
oh well I can't do much about it anyway, I just wonder if other people think I don't know how to ping people with @.
@Snowman Is it ok if we think you don't anyway?
really, it's probably just you're bumping into the 2 pings in one comment, or your browser is choking on the javascript or a security plugin in your browser is kicking it out
user55340
@GlenH7 there is also the "only two " one where there can only be one person you are pinging that then lacks the @
something mundane like that
user41796
Apropos of nothing: Props to @AshleyNunn for dropping this week's humble bundle link in chat recently.
@RobertHarvey I guess I did get it right. Peanuts for me!
17:39
Man, our SCCM client is AWFUL
EVEN FOR @#%@#ing FREE SOFTWARE
@enderland I'm thinking CMMI...no, I have no clue what SCCM is...
system configuration center manager
ohh, you mean MOM
how software is deployed
Oh no n/m, MOM is the old version of a new acronym that's not that one...
to be fair, I've only heard bad things about that windows DC deployment stuff
people tend to prefer non-MS solutions for that I understand
17:42
sudo apt-get
user55340
Mom doesn't work here. Pick up your own messes. In a day or so.
@MichaelT MOM was good, I don't know why people didn't like it or it's replacement...
(no, the replacement wasn't called WIFE)
@JimmyHoffa Kudos on the answer to that guy's question. I really need to get better at this whole "higher-order functions" thing.
17:57
@RobertHarvey aye, that one was more about thinking in terms of types though, encoding rules into your type system
user114359
18:17
We need a vote close reason for Offtopic: please do not post while drunk and please use punctuation. programmers.stackexchange.com/q/272772/22815
that's what the other field is for
Cool, very elementary-school-terms lecture on left and right associativity and recursion in parsers
18:46
^^This is the key. Create a "Standard" and stick to it, don't chop and change every five minutes. Also, this question is better suited to programmers.stackexchange — GPPK 1 min ago
@GPPK Primarily opinion based questions are going to be closed on Programmers, on StackOverflow, and everywhere else on Stack Exchange. Please see Programmers What is On Topicdurron597 1 min ago
19:21
INTERVIEWER: We need someone with mahogany experience.
APPLICANT: I've built dressers and beds with mahogany.
INTERVIEWER: Yes, but we need someone who knows how to build end-tables. We have no interest in dressers or beds.
user41796
@enderland They've got their claws sunk in deep, I take it?
in Duga's Playground, 4 mins ago, by Duga
It's a very broad question (but certainly not off-topic!), however I think it's safe to say that you may find more suitable reactions to it on either programmers.stackexchange.com or dba.stackexchange.comMichal 39 secs ago
I was just thinking about building a game in HTML5 so that I could demonstrate that I know HTML5 and CSS3. Then I remembered that most employers with business application needs will be totally unable to relate. "We don't need games, we need business applications."
I think I need to refine @Duga's filter a bit
user41796
@SimonAndréForsberg We'll find the happy medium between false-positives and missed alerts. Thanks again for working on this.
user114359
19:26
@RobertHarvey I thought building a game was the de facto way to learn anything
@SimonAndréForsberg can you detect links? Links are likely to be positives unless they link to a specific non-meta answers or questions.
@Snowman Not enough jQuery or Angular or [insert favorite data binding technology of the week here].
user41796
@enderland That's a classic end-run maneuver.
@GlenH7 yeah i just forgot to talk with him since I talked to my other two managers...
@amon yup, I can. I think a pure link-detection for Programmers is one of the features in the next version of the filter.
19:29
:20000799 I don't think he knew that, just got lucky - though I'm really glad I got to have the convo I had with my team lead regarding other convos I'd had
user41796
@enderland I'm in a mood, so I'm going to assume malice. :-D
> Peter Gibbons: It's a problem of motivation, all right? Now if I work my ass off and Initech ships a few extra units, I don't see another dime, so where's the motivation? And here's something else, Bob: I have eight different bosses right now.
Bob Slydell: I beg your pardon?
Peter Gibbons: Eight bosses.
Bob Slydell: Eight?
Peter Gibbons: Eight, Bob. So that means that when I make a mistake, I have eight different people coming by to tell me about it. That's my only real motivation is not to be hassled, that and the fear of losing my job. But you know, Bob, that will only make someone work
user41796
@enderland "be patient"
I know. It's just all so funny to me the timing
@enderland It's absurd to have more than 1 boss.
user41796
19:32
Sometimes that's how it works. Props that the current supervisory types are aware that options exist.
The more the merrier!
user41796
@durron597 8's a good number though
@GlenH7 well of course - that's a common enough thing in software dev...
user41796
@enderland I've met a lot that are in complete denial of that option, surprisingly.
@GlenH7 Well, right. You can have 1 boss or 8 bosses but no other numbers work. 8's just one of those magic numbers
user41796
19:34
@durron597 exactly and it just takes a little more jui-jitsu to work them off of each other to your benefit. Smaller numbers of bosses just don't play that way
psr
psr
@durron597 "So can I get that raise now?" "Magic 8-boss says 'Reply hazy, ask again later'".
user41796
When someone posts the compiler output in response to a request for debugger output, you know there's something fundamentally wrong.
user114359
When someone posts a "my code doesn't work" question and shows zero understanding of how code should work, you know there is something fundamentally wrong.
i am always a fan of "my code doesn't work", when they post their source code
and thats it
user41796
I hadn't read Lippert's how to debug small programs yet. There's some hilarious gems in there
19:49
> If your program still has a bug, obtain a rubber duck. Or if a rubber duck is unavailable, get another computer science undergraduate, it’s much the same.
user41796
I nearly literally laughed out loud at that one.
user114359
Rubber duck debugging is an informal term used in software engineering for a method of debugging code. The name is a reference to a story in the book The Pragmatic Programmer in which a programmer would carry around a rubber duck and debug his code by forcing himself to explain it, line-by-line, to the duck. Many other terms exist for this technique, often involving different inanimate objects. Many programmers have had the experience of explaining a programming problem to someone else, possibly even to someone who knows nothing about programming, and then hitting upon the solution in the process...
Man this week is nutszo
user41796
@enderland more has unfolded?
user41796
There, now everyone has to click through all of those histories...
19:56
Yeah... it'd be more of an intergration ETL type position
but less programming and more ETL
user41796
Random link of the day: en.wikipedia.org/wiki/SAP_SE
most large companies run on something like SAP/Oracle/ibm stuff - huge database types of things that an entire company can run with
@GlenH7 That sounds like something that would sap my will to live
Right, hence my hesitation
Pun <you figure it out> intended
user41796
19:59
@durron597 it's notorious for the quirks of the language it uses
user41796
but that said, enterprises simply can't hire enough developers who know SAP.
It's kinda the equivalent of selling your soul for job security and boring work
user41796
although I'm sure some of that is due to needing folks who know how to program in maple sap as opposed to mahogany sap as opposed to cherry sap as opp....
user41796
@enderland The boring part depends upon what the enterprise has put in the framework
@GlenH7 maple sap is good for doing advanced math
user41796
20:01
large, multinationals with huge supply chains can have some interesting work
user41796
@durron597 fermented maple sap is actually pretty good for a whiskey too
@GlenH7 I've also heard vermont maple sap is good on pancakes
I don't know. FML too many decisions
@enderland Close your eyes, scrub your brain, take a deep breath. Once you have successfully blanked out your thoughts, ask yourself "what do I want to do?" Do the first thing that pops into your mind
(That exercise works better when someone else is present to ask the question, but hey.)
I'm from the Workplace. it's "quit"
3
Which... :|
20:08
@enderland this industry is so utterly jacked... hard to find a decent job, hard to find employees, how do you have both of those problems in one industry?? All the reams of people between the want-job folk and want-employees folk are just fucking everything up for everyone.
2
user114359
everyone has unattainable standards
user41796
@Snowman And here I was thinking mine were too low... :-)
user114359
@GlenH7 my employer is very selective about who they hire. But they also have a lot to offer. Job market in my area is actually very healthy so it works out.
@JimmyHoffa The cost of living is so high that only enormous companies can afford to hire people and pay them a decent amount. However, enormous companies also tend to have enormous bureaucracy
user41796
go vote this up, please:
user41796
user114359
vote on what?
@SimonAndréForsberg while you're tuning it, couple crap-throwers from SO slipped through. :( Good that we've got Duga's playground for unfiltered stuff...
What we need is a six-steps-of-separation experiment where people list all the "these people are intelligent and good" and then invite them to list people, with some sort of double blind filtering process
then everyone can just make a list of all the non-terribad people to work with and hiring gets easier
user41796
@Snowman The community promotion ad promoting the new engineering beta site
"you need a reference to join this list! find someone intelligent and convince them you aren't a moron"
20:26
@gnat yup, I am aware of those. I am marking false-negatives in @Duga's Playground by starring them. I have some ideas for how to catch them better
@SimonAndréForsberg to avoid misunderstanding, I don't complain. Your efforts are much appreciated
@gnat No misunderstandings occurred :)
I love seeing double messages, "p.SE!" "no, not P.SE!"
I understand that 1 chat message is better than 0.
@enderland In the 2nd Monitor, that tends to give us something to star (as if we needed that)
Two is good, 1 is bad, 0 is undefined
1 means that no one wanted to correct the chatbot :)
20:31
or that the second wasn't caught by the bot
Yeah I guess
user41796
@enderland Depends. I've taken to referring to the site as Progs when I'm correcting an errant recommendation
@GlenH7 so its your fault! :P
user41796
@Michal - this would get closed on Progs as "Too Broad." — GlenH7 1 hour ago
Someone should make a genetic algorithm to continually refine the filter. If there's a reply comment, the template gets to breed again
20:34
only from a list of users though
@durron597 that someone is not me. But while that someone is at it, that someone should also make a pull request
@durron597 I don't cheat on my dear Code Review with PCG!
@SimonAndréForsberg I was changing the subject to that awesome thread
PCG does have some very interesting challenges every now and then, but at other times it looks more or less like the Kid's Playground of Stack Exchange.
20:43
I should have wrote an example for that GA problem
Every time I encounter a new GOF pattern, I keep thinking "This is a lot of complexity without a correspondingly awesome benefit."
@RobertHarvey new GOF pattern is an oxymoron
Well, new to me.
user41796
@RobertHarvey keep in mind that those were created during the darker days of C++ prior to even the STL being commonplace.
That's good to know. There's a good chance that e.g. the Decorator Pattern has a better solution today.
user41796
20:48
There was a lot of repetitive boilerplate where it made sense to sweep stuff under the pattern rugs.
Reams of boiler plate. No wonder people just parrot this stuff without thinking about it too much.
@GlenH7 which makes it doubly sad that all these years later and most people still haven't improved upon them by, I don't know, creating better solutions to their problems than some canned solutions for problems that are hardly related to theirs 20 years ago
@RobertHarvey oh that. Then probably half of them listed in GOF are new to me either. Long time ago I tried very hard to remember these and couldn't. Dropped this and now load my brain only with those I use in practice, these tend to stick easily
@GlenH7 I haven't written C++ in a really, really long time. Is STL a C++ equivalent of like java.util?
user41796
Put another way, GoF is a set of very intelligent architects who talked with lots of folk. And they kept seeing people complain about the same crap all over the place. So they captured that crud in the concept of patterns
user41796
20:49
@durron597 yes
user41796
As in you don't want to try to write C++ without using it
I wouldn't want to write Java without java.util ;)
@GlenH7 Templates? The only thing similar to those in Java/C# are generics.
user41796
ironically, I didn't get a C++ gig because I didn't have enough experience with mahogany. (aka STL)
user114359
There are some good GOF patterns but some of them are overengineered.
user41796
20:50
@RobertHarvey The STL offers up a lot more than just templates
Oh, didn't know that.
STL is more like a common library - Java has a standard one, C++ for many years hardly did at all, so they put "standard" into the name of their library to try and convince people to standardize on it. It worked more or less, but there's still plenty of non-stl stuff. C++ is like writing Java if some libraries had all their things rooted at ObjectX and others had their shit rooted at ObjectY and then there were a bunch of functions throughout their libraries to convert between the librarys types
user41796
yeah, it was quite ground breaking / revolutionary. One of the key things that shifted folk from writing C code in C++.
user41796
The Standard Template Library (STL) is a software library for the C++ programming language that influenced many parts of the C++ Standard Library. It provides four components called algorithms, containers, functional, and iterators. The STL provides a ready-made set of common classes for C++, such as containers and associative arrays, that can be used with any built-in type and with any user-defined type that supports some elementary operations (such as copying and assignment). STL algorithms are independent of containers, which significantly reduces the complexity of the library. The STL achieves...
how sad that the idea of a standardly used library for their language was "ground breaking" just speaks to how much utterly broken crap C++ people are willing to put up with instead of fixing
user41796
20:53
@JimmyHoffa it was part of the growth of the language
user41796
remember that C++ wasn't backed by some mega corp
user41796
It was basically one guy who said "we can do better than C and the like"
user114359
STL is templates. The "standard library" is everything else.
@GlenH7 neither was perl or python but they had a common core library
user41796
Truthfully, I think he was trying to cram smalltalk upon the rest of us, but he used C's syntax
20:54
Also, while I don't particularly like C++ (I am a Java person), what's really neat is that you can still do C++ on really small systems, with little or no overhead. The STL is not an essential component, just a nice to have, and you can throw it out if you don't need it.
HAH. C++ is so far from smalltalk that's just silly
C++ on a 8-bit microcontroller, no problem
user41796
@JimmyHoffa some of the conceptual ideas came out of there, IIRC. I may have misplaced it's actual inspiration
@rolfl yeah but then you may as well write C and have a language that better fits the paradigm you'll be coding in anyway if you're working that embedded
user114359
@rolfl part of that is because of the linker model, not so much the C++ language itself. Static linking will generally omit dead code, shrinking the size of your executable which is useful for low-footprint programs.
user114359
20:55
Same is true for C too
@GlenH7 people often attribute "inheritance" and "objects" to small talk and then say C++ must have gotten those ideas from it. Truth be told the type systems aren't very similar between the two languages
That is all true, but the distinction between the language and the libraries is much clearer than in languages like Java, where the language definition depends on the libraries...
Consider the Java structure: for (String s : args) { ....
@rolfl yes, but some would argue you shouldn't try make decisions for such a specialist niche that causes so many problems for your broader purpose - if C++ truly intends to be "general purpose".
Java doesn't intend to be that embedded so there's no incongruence in them deciding to tie their language to their library when that's the rational decision for a high level general purpose language
(not that Java doesn't aim for embedded, just not at the microcontroller level quite the same as C/C++)
user114359
I thought Java was intended for embedded originally, but ended up dominating web servers instead
user114359
I thought the original marketese for it involved running everywhere, like a 90s version of IoT
21:01
@Snowman it was intended for cross platform but even at that I don't know how much it was intended to be used for tiny level. That said I understand it does still have some libraries and runtime specialized to run at the embedded level. Though all of the decisions throughout the language make it a very hard sell to claim it wasn't intending to be a general purpose language first and foremost. I think the cross-platform thing was a side effect of their intention to have something
that was general purpose and high level
user41796
Simula, not smalltalk. My bad. stroustrup.com/hopl2.pdf
Simula; wow, that goes back a ways.
Excel gurus: is there a keystroke for copy cell contents (as opposed to copy entire cell)
user41796
@RobertHarvey 1979 started the trip fantastic for what is now known as C++
user41796
And I don't think AT&T backed Bjarne in the way that MS or Sun did with C# and Java.
21:09
@durron597 you can double click into the cell and copy it? is that not what you want?
user41796
it was more of having the freedom to give it a go
without putting my hands on the mouse
you might be able to set the default paste mode to be special
the equivalent of F2 -> Shift + Home -> Ctrl + C
user41796
@durron597 I have always used the mouse to click into the formula bar and copy from there
21:12
I should ask my boss at Employer^. Excel and being a jerk were the only things he was good at
7
A: How to copy the contents only of an Excel cell minus the formatting?

jcrawfordorYou can use Paste Special (in the edit menu) to specify exactly what should and should not be pasted. If you have Excel 2007 or later, pasting normally (Ctrl+V) also works; a dropdown box then appears with options to paste without formatting, etc. You can then click the menu or press Ctrl to op...

@rolfl This is a copy issue not a paste issue
key point being, after pasting, you can press the ctrl key again to get through the paste-special options.
I'm trying to copy a cell from one sheet into the Filter search box of another sheet
which doesn't accept the "cell" data flavor
Well, don't do that, then.... didn't you know, microsoft only makes possible the things you need to do. If it is not possible, then you did not need to do it anyway.
21:16
write a short keyboard macro, record it, and assign it to a shortcut
and by write I mean ht the record button and let it do it for you
@rolfl s/microsoft only makes/Microsoft and Apple only make/ (gotta make sure to fix the grammar AND the capitalization)
@rolfl people joke about these things all the time, and then speak despisingly of all kinds of language features that are "foot guns" -> If <corp/tech I dislike> doesn't allow it, you didn't need to do it, amiright? sarcasm sarcasm ? and then they follow it up with "dynamic typing is just wrong, nobody should do that!" or "multiple inheritance just causes problems, no one should do that!"
'twould be nice if Excel boxes in formula editors also had right click menus (especially for cut/copy/paste)
The reality is that the companies are likely purposefully curtailing the abilities of their technology in a tradeoff trying to avoid people accidentally causing lots of problems at the risk of removing their ability to do something that may be useful
especially since arrow keys edit the text instead of moving the cursor.
21:22
the goal being to help those who use their tech do good work, tradeoffs always make someone unhappy.
Ooops... catching up... Yeah, companies compromize between features supported, and user confusion.
I have done that before too. (or been asked to do it).
user41796
@rolfl Always gotta prioritize user confusion.
@rolfl we all do, it's good practice- an important part of API design.
If we weren't worried about making people capable of doing dangerous things we'd just do away with public and private accessors etc
@rolfl I interviewed with Employer^^ a couple of months ago, though we couldn't come to an agreement... One of the reasons I interviewed a second time was I wanted to know how he managed to keep five software developers busy. The answer was that he says "yes" to every feature request a customer makes. His code base is now two million lines of code and 92 separate C# and VB projects.
I mean, that's why vi does not support playing doom, whereas emacs probably does ;-)
21:31
so that line of argument against a company or tech really holds no water. It's just a populist line of garbage.
@JimmyHoffa Of course it is, but it's also fun (sometimes)
I work for the king of companies that does that - IBM.
You may have the features of your choice, so long as they are blue.
@rolfl ah, I'm sorry. I've yet to run into a single soul who enjoyed working for IBM in the last 20 years...
I got it lucky. I have a lot of fun, actually.
I've heard great things about the 70s and 80s IBM, but you're the first to claim to have a decent space in any modern incarnation of it. Though not surprising, huge companies are huge. The worst huge company still has a good team somewhere just by sheer odds.
In a company of 400,000+ people, if you can't find something interesting to do, it's not their fault...
21:35
@rolfl That's because you need 390,000 drones.
In fairness, IBM has this thing called an IBM fellow, a person who is paid lots, and given money to make good ideas happen.
Yeah; they're in the 10,000.
I've always found the work in smaller companies far more interesting.
I happen to be working in one of the 'innovation teams' that's been put together in by a fellow.
@rolfl yes, this I've heard of. Different story than the majority of their employees. Like Microsofts "Partners" if I recall the term correctly
user41796
@JimmyHoffa Fellows usually have to work their way up through the ranks. Booch is an oddball since he was picked up when IBM bought Rationale
user41796
21:37
Ditto with Gamma (at least I think it was Gamma they picked up with Rational as well)
Many fellows are acquired.... (sort of becomes a privilidge of being acquired).
"hey we want to keep you but dont have a place, here, take some money and a team"
I worked for Algorithmics when they were acquired, then Neil Bartlett became a fellow.
user41796
@rolfl one of the guys I worked with received his in 2009. I just looked him up.
Jeff Jonas, from G2 when they were acquired.
00:00 - 22:0022:00 - 00:00

« first day (1621 days earlier)      last day (3666 days later) »