« first day (1127 days earlier)      last day (3866 days later) » 

6:48 AM
@GlenH7 do you plan to update analysis for clarified hotness formula? if yes, please note that current analysis most likely still applies to per site hot list which is same crap, only the damage is less prominent - ie worth keeping at very least a link to current analysis revision
 
 
2 hours later…
8:43 AM
I can understand if install and uninstall programs made by non-Microsoft companies don't work. What blows my mind everytime is the fact that MICROSOFT SQL Server freezes upon uninstall.
Find a registry key that you don't like? BLAST IT TO HELL! You OWN the operating system afterall
 
 
1 hour later…
10:11 AM
Morning all
 
10:56 AM
@fredley haha this is fun
 
user41796
11:10 AM
@gnat Yes, I'll re-run the base numbers. Doesn't make sense to re-run the comparisons of the suggestions though since they've been rejected. There may be some patterns that come across with the new numbers to suggest other ways of tamping down top bar collider values.
 
@GlenH7 sounds good! do you consider also keeping a part of (or at least a link to) a current analysis? I wouldn't want it to get buried too deep as it might turn out helpful to deal with per-site lists
 
11:58 AM
good food for thought...
99
Q: The problem with extrinsic motivation

Jon EricsonAs a preface, I've stopped using Stack Overflow after participating in the beta and using the system for a while. I'm not trying to stir up controversy or ruffle feathers, though that seems likely to happen. Rather this is a postmortem of one user's experience with Stack Overflow. I've always ...

 
@gnat my problem with rep is it kills all extrinsic motivation to answer questions which are on not frequented tags
posting 2 answers on C# or C++ or java get you the same rep as 10x as many (or more) in less popular tags
 
12:51 PM
I'm just going to drop this link in here and then sit back and hope something interesting and/or amusing erupts.
Best coding practices are a set of informal rules that the software development community has learned over time to improve the quality of applications and simplify their maintenance. They can be broken into many levels based on the coding language, the platform, the target environment and so forth. Using best practices for a given situation greatly reduces the probability of introducing errors into your applications, regardless of which software development model is being used to create that application. There are standards that originated from the intensive study of industry experts who...
 
@ThomasOwens my thoughts on the importance of comments have changed significantly over the past few years
 
1:07 PM
shameless self bump...
0
Q: Rebuild IBM Portal embedded Derby database without a full reinstall

maple_shaftAfter exhausting all conceivable options over a matter of weeks, and after the drudgery of the back and forth with subpar IBM support, I have come to the conclusion that the only explanation for why my specific development environment fails to run a custom theme where other environments have no p...

Go home IBM... your drunk
 
user41796
@ThomasOwens I'm sure that article infringes upon some consultancy's intellectual property rights. Because there can be only one set of best practices.
 
user41796
@maple_shaft True story: TJ Watson Jr was a roaring alcoholic when he was younger. Shaped IBM's anti-booze policy for decades
 
user41796
@gnat I have the analysis built out within a spreadsheet. It's just as easy for me to copy the data portion over to a new tab and work from there given the clarification on the rules. I still don't believe that the time decay is working as they expect it to / how it's advertised.
 
user41796
But I'll verify that and post a clarifying comment to the answer with the current formula
 
user41796
Wikipedia says my anecdote about TJ Watson and alcohol is a little off base. That article claims it was Sr. who had the issue and then banned the whole company.
 
1:20 PM
@GlenH7 working hard on friday eh? :)
 
user41796
@enderland As always. I have found memories of working at Big Blue. And maple shaft's comment brought back an anecdote I had remembered.
 
hmmm I wonder what the best way to handle a database connection class is
if I should connect/disconnect whenever I need to, or keep a standing connection at all times the app is open?
 
user41796
read the documentation on the connection class. For some types of connections, windows will create a pool of connections for you
 
user41796
and then it will keep the connections open for you. So the driver is handling the optimization that you're looking for by trying to balance when to open & close
 
hmmm, I've got it somewhat wrapped into a class I wrote which goes in/out of scope (and presumably cleans up any database connections)
I suspect VBA is too dumb to do that, too, unfortunately
 
user41796
1:28 PM
It's not at the VBA layer, it's at the ODBC / whatever type of DB connection layer instead.
 
@GlenH7 oh gotcha. hm.
 
user41796
for example, with SQL Server and ADO connections: msdn.microsoft.com/en-us/library/8xx3tyca.aspx
 
user41796
AFAIK, ADO and the other MS connection technologies have some sort of pooling for most / all of the major DB providers. This is the article for Oracle: msdn.microsoft.com/en-us/library/ms254502.aspx
 
yeah, I think that's what probably is happening
When the application calls Close on the connection, the pooler returns it to the pooled set of active connections instead of closing
I'm using ADO and SQL server
I'm curious how cleanup happens when my wrapper class goes out of scope
this article seems to suggest it doesn't have to "reconnect"
 
user41796
IIRC, There is some funniness regarding the connection string and credentials, but it should be called out in that article
 
1:31 PM
which honestly is pretty awesome
I'll always be connecting using the same connection string fortunately :-)
it looks like it adds a separate connection to the pool if you use diff string/credentials
this is actually pretty neat. I did not expect it to work like this, though, in retrospect it makes complete sense
I wonder when the connection pool is actually cleared? or when that happens?
 
user41796
It used to be that connecting to the DB was very expensive. So people would open the connection and leave it open. But then you had issues with the connection being closed on the SQL Server, and developers not verifying the connection was still okay before starting to use it again. So this was MS's way of resolving the cost of connection issue to fix bad programming behavior
 
user41796
I believe there are timeouts built into the pool connections. I don't know if the pool persists beyond the application's lifespan.
 
this actually makes a lot of sense, tbh, I've always been curious why some of my queries are so much faster - I bet it's because it's just finding the existing connection rather than creating a new one
 
user41796
that could certainly be part of the issue, yes. First connection is typically pretty expensive.
 
it looks like the pooling happens not at the application level but the ODBC level so I'd guess it persists beyond application lifetime
 
user41796
1:38 PM
Sounds like it yes. Which is kind of crazy in a way. Two apps could theoretically be sharing the same pool.
 
this makes my program architecture way nicer
at least this element of it
trying to write apps to be maintainable by non-devs is obnoxiously hard
 
user41796
@enderland um, yeah. No thanks. eek
 
@GlenH7 fortunately it's reasonably small scale
man I wish I had known about SE when I was in undergrad
I'm pretty sure I would have learned 10x as much
 
user41796
depending upon how long you've been out of Uni, it wouldn't have mattered
 
user41796
SO kicked off in 2008 or so. P.SE kicked off in ~2010.
 
1:43 PM
so let's amend that to, "wish I had known about and had it be around" :)
 
user41796
yeah... I hear you on that one. I'm kind of surprised Usenet is alive and kicking though
 
user55340
@enderland Man... I wish SE was around when I was an undergrad... I had usenet instead...
 
user55340
2:01 PM
That, and swedish programmers. Let me explain.. I was a 'wizard' on a 'mud' (LP style, object oriented message passing Cish language). It was in Sweden. Sometimes, when I had a bug in my homework, I'd upload it to the mud and talk it through with the other wizards (who were programmers themselves most of the time).
 
> Original Author
> Modifications
> Authors who modified code with a description on why it was modified.
yeah, those are great comments to have when you have all your code in source control, that shit isn't annoying at all.
> Questions to ask yourself while coding:
> Is this a small function which will never be replaced ?
> Will there be any modifications to this function ?
If you're asking either of those questions, you're still a junior
 
user55340
@JimmyHoffa Be reasonable: No, Yes.
 
@MichaelT Which is why you never need to ask
 
@ThomasOwens What do you think Wikipedia would say if I just took a hatchet to that entire article removing everything that looked like a suggestion, only leaving a description of the concept of "Best Practices"
 
user55340
2:10 PM
@JimmyHoffa They'd get rather annoyed.
 
user55340
Start with peppering it with [citation needed]...
 
@MichaelT Should I care when it's totally full of crap and I'm actually the type of knowledgeable person they want information in their articles from
@MichaelT That's actually a great idea
However it's a pretty horrible idea also
because you can find citations for everything they mention
They are however all highly controversial
 
user55340
Quite frankly, I take @author stuff out when I see it. I don't care who wrote it. If I want to, I look at git blame to see whats where.
 
In fact they have a citation of one thing right here: docforge.com/wiki/Best_practices
@MichaelT Same.
 
user55340
Dig into that docforge a bit, and you'll see the documentation page has no such suggestions - docforge.com/wiki/Code_documentation
 
2:14 PM
@MichaelT Yeah, it actually doesn't, should I blow away the citation?
 
user55340
I'd start with poking the talk page, say what changes you intend to make, that the suggestions are original research (use that term) and is not backed up by the citation and is at best, controversial.
 
Actually the citation says the opposite of it
 
user55340
Check back occasionally, if there's no response or a consensus in your favor, make the change.
 
He uses that as a citation for "Always make logic as simple as possible", but the page he cites that from says over-simplifying can be as risky as under-simplifying
I've tried multiple times to correct something which is so utterly wrong in wikipedia, that really pisses me off:
 
user55340
The key to making (signficant) changes within wikipedia is to know how to play within their rules.
 
2:17 PM
Golden Ale.
 
user55340
There's a template for "says who" or something to that effect. I'd drop that after the "one commenting practice"
 
They say golden ale is pale ale, which is dead fucking wrong. Go find a pale ale, then have a duvel, or delerium tremens, or countless other actual "golden" ales
 
user55340
> Example: A variable for taking in weight as a parameter for a truck can be named TrkWeight or TruckWeight, with TruckWeight being the more preferable one, since it is instantly recognisable. See CamelCase naming of variables.
 
user55340
Notice the caps on the variable names.
 
user55340
That section that is most painful was all done by one person - en.wikipedia.org/w/…
 
user55340
2:21 PM
Given that, I'd suggest going to the talk page and suggesting removing those blocks as unreferenced, and original research that are at best personal opinion and not backed up by any external sources.
 
@ThomasOwens you're a jerk. Now I'm genuinely pondering learning the Wikipedia platform. What a crock.
Ah hah, I found the source of most of the information, a hand full of edits from this one guy en.wikipedia.org/w/…
I should toss in there "You should always make all of your code OO, and it is a best practice to do web development in PHP, or if your time line is short Ruby is acceptable."
 
user55340
2:43 PM
I'll poke at removing those sections when I get home this evening or this weekend. Reasonably established minor editor that people won't out of hand accuse me of being a driveby editor.
 
@MichaelT I've driveby edited golden ale definition corrections multiple times, people won't have it. Grumble. Duvel is NOT a pale ale, nor is any other belgian strong golden ale a "pale ale" of any sort.
 
 
2 hours later…
psr
4:54 PM
@JimmyHoffa No, they don't. They don't want experts (they have no way to tell an expert from a fraud), they want citations. Answer a "best practices" question on Programmers, then cite that and you might have a shot.
 
Also, it turns out this site exists...
 
psr
@JimmyHoffa I found that when I read the Time comic and just wasn't going to put that much of my time in.
 
 
2 hours later…
user55340
7:15 PM
@WorldEngineer Thank you for duping that question... I realized the dup after I had the cv in there for off topic.
 
@JimmyHoffa inb4TheWorkplace
 
@enderland ...but nobody cares about the work place, that's where SE sends questions to die right?
 
@JimmyHoffa that's where Programmers sends questions to die
 
@enderland the amount of spongebob I've had to see because my kid is what's not cool...
@MichaelT I hate duping questions that are unacceptable..
 
7:43 PM
@JimmyHoffa lol, I didn't even know that was - just one of the images for "not cool" :)
 
user55340
@JimmyHoffa I keep sitting on the fence on that. It hurts. On one hand, it sets up another sign post to the proper questions that have been answered (and closed). On the other hand, it doesn't put up a 'stop' sign.
 
@MichaelT While I agree, I think that the perception by most isn't "The question it's duped of is closed and off topic" but "The question was asked before and they duped my question to it, so it's on topic", the last part is what makes the decision up for me.
Not the lack of a stop sign, but what appears to many as a Go sign
 
do you guys get this sort of crap here too?
When one realizes that he is in the unemployment line because of being a jerk he does not need a long answer to his own predicament. — Theodore A. Jones 3 mins ago
(after a 10 word answer)
 
user55340
@JimmyHoffa Those that would interpret it as a go rarely look enough to see a stop or a go.
 
@MichaelT I don't know that I buy that. Programmers are all skimmers, comes with the territory, but that doesn't mean if the question was closed they wouldn't get the picture
 
user55340
7:51 PM
Now, what could be done is taking those dup chains on education / college and have a mod reopen / reclose them to the modern close reason.
 
user55340
So rather than 'not constructive' or 'not a real question' or 'too localized' they are 'off topic career advice'
 
@MichaelT Yeah, but that flies in the face of "Don't hold old content to current content standards"
Many of those old ones aren't closed but locked
or just cobwebbed with hopes of never being reawoken
 
user55340
they're already closed. This doesn't change state of the question but rather helps modern readers see the modern reason when they do look.
 
@enderland you mean snarky jerks? I can't seem to find a place without them on the nteret...
 
user55340
If they do look, seeing a "not constructive" isn't a 'stop' sign that all questions of this category are off topic.
 
7:56 PM
It's snowing here today.
 
user55340
@JimmyHoffa And that said, we often hold old questions to modern standards - closing old musty things that we don't want to be a new user honeypot.
 
@MichaelT Which I appreciate; though we're not supposed to... I suppose it's one of those if-you-don't-tell-I-won't-tell situations
 
user55340
If it gets activity, its subject to modern standards.
 
I wish that rule weren't as it is though. Given the contrast between the old standards and current standards, it just seems unreasonable to leave a lot of that stuff in place and yet somehow tell people it's not ok to do precisely that stuff
 
user55340
Sam's editing again... fortunately the question is already protected.
 
8:09 PM
0
Q: What does the term "hard coding" or "hard coded" actually mean?

BasI watched a interview with Bill Gates about ctrl+alt+delete, and in a sentence he said "hard coded in the hardware". But i don't know what this term means. Can someone explain this to me? Thanks in advance.

 
user55340
How much research have you done? A quick search of google brings up wikipedia hard coding and wiktionary hard code as the top two results. — MichaelT 10 secs ago
 
This is just really interesting. It genuinely (to my eyes) is a good question. The guy is so far below the rungs he should be to be dealing with anything around here (likely a kid) if he's not familiar with that term. Yet he put together a quality question.
Ok, I suppose it is a bit low on research...
 
user55340
17, netherlands.
 
aaaaaaaand its gone
 
> This question was voluntarily removed by its author.
Cool.
 
user55340
8:12 PM
Probably for the best... I had just cast a cv on it.
 
@MichaelT if only you could undo your downvote on the deleted question
 
user55340
Not my downvote.
 
I was with you on this one @JimmyHoffa it could have been edited to be better
 
user55340
I've been casting upvotes and finally got my sportsman badge!
 
8:12 PM
I'm just surprised how good he is at putting together questions when he is sooo far out of his depth
 
user55340
He's got a few questions on SO.
 
8:25 PM
HI all
 
@MichaelT wow. nice.
 
@jozefg Good answer on that math for haskell question. I figured you'd pop in with details of the different areas to really poke at
(now I've got to go find the "Lambda Cube" because that's new to me...)
 
@JimmyHoffa Thanks :) I just kinda pop up when someone asks about type theory, it makes my lambda senses tingle.
 
user55340
Any python masters want to poke at this? @WorldEngineer ?
 
user55340
0
Q: Why doesn't Python just establish a format for declaring tab widths?

IndoleringI've read the flame wars over the use of spaces and tabs. When working with any markup language (when scope isn't very important and pressing space 4 times is a PITA), I tend to minimize the tab width to just a space or two but I fluidly change to four full tabs for my other programming work. It...

 
8:32 PM
@jozefg Ok you said Haskell is Lambda Fw; my reading of the lambda cube I would say that's where it wants to be, and maybe gets relatively close with lots of GHC extensions, but due to the fact that they won't sacrifice turing completeness they can't be totality checked and therefore are really Lambda w. Somewhere up the axis towards Fw with GHC extensions perhaps
am I misunderstanding the lambda cube?
 
user55340
See, when I hear people talking about lambda cube... I think Time Cube...
 
user55340
Time Cube is a website created by Gene Ray, also known as Otis Eugene Ray, in 1997 John C. Dvorak wrote in PC Magazine that "Metasites that track crackpot sites often say this is the number one nutty site." Concept The website is mostly text written in centered, multi-colored 30-point type in a single vertical column. The following quotation from the TimeCube.com website illustrates a recurring theme from Gene Ray's ideas: Ray has wagered $10,000 that his theories cannot be proven wrong. John C. Dvorak of PC Magazine characterized the site's content as "endless blather". Public reacti...
 
Lambda Fw is type operators + system F, both of which are in Haskell 98. Dependent types are in Lambda P
 
@jozefg O I misread, I got the P and F mixed up. Carry on.
@MichaelT Which was awesome. I tried to get people to read this many years ago just to make them think I was crazy. (Ok I am but beyond the point)
 
@JimmyHoffa Plus you can have turing completeness + dependent types, it's just unsound and screws type inference royally :)
Unsound in the sense of it's possible to inhabit arbitrary types, which is bad news when a type inhabitant is a proof of a theorem
 
8:44 PM
@jozefg Which is why you can't have it totality checked
To my eyes totality checking is the pinnacle of the purpose of dependent types, but totality checking requires halting thus, no good dependent typing with turing completeness
(until some folks wayyyy smarter than I figure out how)
 
 
1 hour later…
user55340
9:55 PM
 
10:13 PM
@MichaelT great video
 
user55340
forwarded to me by a cow-orker.
 
I died a little when I wanted to check my phone 30 seconds in
I definitley think about this when I'm fairly distracted while working
maybe I should unplug more :)
though to be fair. it's 10 min before I leave on Friday
 

« first day (1127 days earlier)      last day (3866 days later) »