The Nineteenth Byte

12:08 AM

Est-ce que quelqu'un aimerait de bavarder dans une conversation en francais dans <<le dix-neuvieme byte>>? chat.stackexchange.com/rooms/43239/le-dix-neuvieme-byte

(my french is probably too bad for anyone to understand lol)

HyperNeutrino

12:24 AM

yay I posted my first APL answer here

ATaco

@HyperNeutrino Have a TIO CMC Userscript.

HyperNeutrino

@ATaco :D :D :D thanks! :D

did someone downvote my neutrino bot post or was that SE's autodownvote thingy

Downgoat

12:56 AM

so our (HS) physics textbook has an experiment that apparently that lists (radioactive) U-235 in materials >_<

ATaco

1:14 AM

I've used 235 in an experiment, although Radon-222 was more common.

HyperNeutrino

uh well then

I wish we got to do something interesting

maybe throw some Cesium into water? idk

ATaco

Our physics teacher said something along the lines of "I'm old, it doesn't matter if I get cancer"

HyperNeutrino

instead we get to shine ray boxes at mirrors to observe the fact that incident rays have the same angle as reflected rays, which I knew when I was in like grade 2

@ATaco wtf lol

ATaco

It was almost definitely just a light hearted joke, nobody was exposed to either substance for long enough to do anything.

HyperNeutrino

ok good :P

Downgoat

1:16 AM

@ATaco O_o where did your HS get U-235

are they even allowed to handle radioactive substance

ATaco

Australia.

We're one of the biggest exporters of Uranium.

Downgoat

oh, I guess all that radioactive stuff is what makes spider so giant in australia

ATaco

(But I have no idea why it was legal)

Maltysen

@Downgoat uranium like that isn't too dangerous

im sure u can buy some online

HyperNeutrino

idk I feel like HS shouldn't have radioactive materials??? what if a psycho makes a nuke in the school? but yeah I don't think our HS carries radioactive hazmat; at least I don't see any long-term health problem or radioactive material symbols on storage doors

Downgoat

1:18 AM

@Maltysen TIL you can buy your way onto FBI watchlist

ASCII-only

@HyperNeutrino Uh obviously the high school doesn't have enough to make a nuke...

ATaco

You seem to have a misunderstanding of the difference between a radio-active substance and the tools required to build a nuke.

El'endia Starman

@HyperNeutrino Look into nuke-making sometime. It's not that easy, by a long shot.

HyperNeutrino

@ASCII-only well yeah I was joking :P

Downgoat

amazon has uranium but no prime :(

El'endia Starman

1:18 AM

There are also fascinating technical details.

HyperNeutrino

oo I should research that. preferably while at school so the school board will look at my search records suspiciously

isn't it 236 or something?

ATaco

And we successfully determined that a fridge magnet was powerful enough to block 100% of radiation that was detectable by our Geiger.

HyperNeutrino

uh lol

ASCII-only

236 has like 0 half-life IIRC?

El'endia Starman

The main uranium isotopes are 235 and 238.

HyperNeutrino

1:19 AM

yeah actually IIRC they throw a bunch of neutrons at 235 and it becomes 236 and chain-reacts to blow up or something like that?

Downgoat

I think in like the 30s they used to use Uranium plates/silverware before they new it was radioactive and everyone died, but it wasn't like kaboom dead, so probably not too bad

ASCII-only

@HyperNeutrino you don't increase in mass by ejecting neutrons

Downgoat

@HyperNeutrino I mean I don't think it really ever 'becomes' 236 it just splits immediately

Maltysen

yeah i think if u have it in a school, just like sotre it in like lead-linded boxes or something

HyperNeutrino

well yeah true

Downgoat

1:20 AM

:| isn't lead also bad

ATaco

@Maltysen Or fridge magnets apparently.

HyperNeutrino

@Downgoat Everyone Dies TM

Maltysen

@ATaco lol

ASCII-only

@Downgoat I don't think it ever becomes U-236 at all

El'endia Starman

@Downgoat Y'know, I think that if you narrated a How It's Made kind of TV show, that'd probably be quite entertaining.

Maltysen

1:21 AM

its just a goat walking around the factory with a go pro mounted on its head

ATaco

$^{235}U$ decays into $^{231}Th$

To make $^{235}U$ into $^{236}U$ You'd need to bombard it with neutrons, and I'm not sure how easy that is.

El'endia Starman

@Maltysen "And here you see radioactive water contaminated by uranium. *bump* Uh, let's go look at another part of the factory!"

HyperNeutrino

@ATaco easy, throw them like baseballs

Downgoat

@Maltysen what do you mean 'uranium-235 is not tin can' :P

ATaco

Have some information in the formation of $^{236}U$

Downgoat

1:24 AM

> Uranium-236 is an isotope of uranium that is neither fissile with thermal neutrons, nor very good fertile material, but is generally considered a nuisance and long-lived radioactive waste.

HyperNeutrino

oh so that's what radioactive waste is

just when 235 refuses to fission (that's a verb?)

ASCII-only

@HyperNeutrino Part of it

ATaco

to fiss..?

Downgoat

@HyperNeutrino word you're looking for is fude

HyperNeutrino

huh

that is a weird word

Maltysen

1:26 AM

@HyperNeutrino also just everything else that is around any of the radioactive stuff

ATaco

you fuse to make fusion, do you not fiss to make fission?

ASCII-only

@HyperNeutrino Correct, although I'd use split

ATaco

Apparently "fiss" is pronounced "Fish"

Downgoat

btw is the Nuclear Shell model still accepted at the current time?

ASCII-only

@Downgoat ~~accepted? no. used? yes~~ yes iirc

ATaco

1:28 AM

(Not the electron shell model, by the way)

Details~

Downgoat

@ASCII-only wait really wikipedia article say otherwise

ASCII-only

@Downgoat 1. wait really 2. i said iirc :P

Downgoat

btw can u work on vsl today

ASCII-only

yes

Downgoat

:D ok joining C9 then

Christopher 2EZ 4RTZ

1:48 AM

@Dennis mod help plz codegolf.stackexchange.com/questions/43150/… can you nuke comments?

90% are not applicable anymore

@HyperNeutrino Because you helped me randomly lol I had large items but not a smexy particle

Also hi o/

ASCII-only

@Christopher2EZ4RTZ How is it 12kb if that code is 54kb

Anonymous

@Christopher2EZ4RTZ Raise a custom flag on the post; more effective than pinging a single mod in chat

Christopher 2EZ 4RTZ

@ASCII-only Because golfed version vs version I used to score

Different since I added a few features so I could auto score my stuff

ASCII-only

@Christopher2EZ4RTZ pls include golfed version too

ATaco

Sorry Chrissy, Flagged as NAA.

Christopher 2EZ 4RTZ

1:53 AM

wat?

ATaco

The golfed version is not in the answer, even if it's zipped up or whatever.

ASCII-only

@Christopher2EZ4RTZ large items?

Christopher 2EZ 4RTZ

I can't even fit it in

ATaco

Then remove the ungolfed version?

Christopher 2EZ 4RTZ

@ASCII-only A human, a taco, a down goat

ATaco

1:55 AM

You know, replace the optional one with the non-optional one?

Christopher 2EZ 4RTZ

Ohh

I actually messed up majorly when counting bytes

ASCII-only

...

Christopher 2EZ 4RTZ

*and also deleted the golfed version

ATaco

>.<

>.<

HyperNeutrino

> helped me randomly

i do not have any memory of this huh?

ATaco

1:56 AM

AFK.

Christopher 2EZ 4RTZ

@HyperNeutrino Geez never thought I would have to explain being thankful XD

HyperNeutrino

lol

Christopher 2EZ 4RTZ

I mean I can give you a footnote saying you where horrible and drove the community away so to prove how great I was I proceeded to call you mario because you got 1 uped

Wtf did I just type

Maltysen

@Christopher2EZ4RTZ the last thing was from an xkcd

i remeber that one

Christopher 2EZ 4RTZ

good job

HyperNeutrino

2:01 AM

@Christopher2EZ4RTZ I mean you could say that and I wouldn't really care until people started to judge me lol

Christopher 2EZ 4RTZ

judgmental look of disapproval ಠ__ಠ

Pavel

smbc-comics.com/comic/algebra

HyperNeutrino

@Christopher2EZ4RTZ ಠ_ಠ

Christopher 2EZ 4RTZ

I cast a vote to re open and opened a challenge, I then realized it was a dupe (that I just reopened it from) and cast the first dupe vote

Downgoat

> they are not useful in the real world

Pavel

2:05 AM

@ATaco please fixಠ__ಠ bug

Downgoat

yes phones, cars, electronics, science are definetly useless

Christopher 2EZ 4RTZ

How could a goat even use a touchscreen?

Pavel

@Downgoat Presumably if that's the kind of thing you want to know about you could take an Alg/Calc course in Uni.

Christopher 2EZ 4RTZ

I am in AP calc

regret

Pavel

I am also in AP calc
not regret, quite enjoyable

Better than APUSH, at least.

Christopher 2EZ 4RTZ

2:12 AM

Why would you take that

@Pavel AP calc is fun but hard af

Downgoat

@Pavel I disagree, HS should provide you with the basic knoledge of fundamental ideas (e.g. Algebra, calculus, nuclear theory, etc.) so when you walk into uni you can comprehend the higher level concepts thrown at you

Pavel

Well, yeah, it's obviously satire >_<

Christopher 2EZ 4RTZ

Take it ez on the goat

Big ball of mud

A big ball of mud is a software system that lacks a perceivable architecture. Although undesirable from a software engineering point of view, such systems are common in practice due to business pressures, developer turnover and code entropy. They are a type of design anti-pattern. == In computer programs == The term was popularized in Brian Foote and Joseph Yoder's 1997 paper of the same name, which defines the term: A Big Ball of Mud is a haphazardly structured, sprawling, sloppy, duct-tape-and-baling-wire, spaghetti-code jungle. These systems show unmistakable signs of unregulated growth, and...

Most honest name

ATaco

> A Big Ball of Mud is a haphazardly structured, sprawling, sloppy, duct-tape-and-baling-wire, spaghetti-code jungle.

I'm pretty sure I wrote that code :P

Helka Homba

Any for contact or codenames? chat.stackexchange.com/rooms/53490/contact chat.stackexchange.com/rooms/59120/codenames

ATaco

2:26 AM

> APL is like a beautiful diamond – flawless, beautifully symmetrical. But you can't add anything to it. If you try to glue on another diamond, you don't get a bigger diamond. Lisp is like a ball of mud. Add more and it's still a ball of mud – it still looks like Lisp.

I'm always keen for Contact.

Helka Homba

Your legoness makes you look like a waffle-taco.

ATaco

I've heard that a fair few times.

Helka Homba

A waffle iron might work well as a tortilla press...

ATaco

I found the Wikipedia article for C's libraries!

Pavel

@TuxCopter I made a pull request for Sakura that should make it usable without cding into the interpretter's directory, can you take a look?

ATaco

2:39 AM

I wonder what Java does if you try to make a stack with more than 2^32-1 elements...

(Besides cry because that's more than 4GB)

Anonymous

@ATaco Crash and burn, unless you're using 64-bit Java, which you should be

Pavel

So you're saying it won't crash and burn with 64bit java

ATaco

At which point I'd want to try to make more than 2^64-1 elements.

.size returns an int, however Stack doesn't use a typical array format, and size is generated on call.

Pavel

int[] foo = new int[Integer.MAX_VALUE]

ATaco

That there isn't a stack

Pavel

2:46 AM

Yes

Anonymous

@Pavel That's only ~4 GiB though

Pavel

@Mego long[] foo = new long[Long.MAX_VALUE]

I don't even know if you're allowed to do that

ATaco

@Mego Actually, that should theoretically be sizeof(int)*4 GiB

Pavel

In Java, sizeof(int) is 4 chars.

So 16GiB?

ATaco

I suppose.

Ethan

2:54 AM

Off topic, but by chance does anyone know where one or people in general can make suggestions for functionality/design on the stackexchange network as a whole?

TuxCopter

@Pavel sure

Pavel

ಠ_ಠ Rainbow Six detected my computer as running Windows Vista. (It actually has 10)

Anonymous

@ATaco Nope. Integer.MAX_VALUE is 2**31-1. Thus, it would be 32*2**31-1 = 2**36-1 = 8 GiB. My math was off, but so is yours.

Anonymous

(also a bit extra memory for the dynamic allocation)

Anonymous

Ethan

2:57 AM

For example, users with a certain amount of reputation are given the ability to look over and approve/reject edits to new questions. Now it doesn't do this now but perhaps it would be more efficient if the questions given to people to approve/reject were ordered by the reputation of the user making the correction, as empirically users with more reputation are probably less likely to make improper edits, vandalization edits etc. now of course you want everything to be reviewed,

ASCII-only

Rip

Really need to chkdsk my laptop soon

Ethan

but by prioritizing you get more done, like in the same way police prioritize which crimes they go after to achieve less crime

ASCII-only

Also back it up

ATaco

@Mego Is that not GibiBits, not Gibibytes?

New Main Posts

0

Q: Fill in the Blanks, Please!

(No, not this nor any of these) Given a string and a list of strings, fill in the all blanks in the input string with corresponding strings. Input/Output The input string contains only alphabetic characters, spaces, and underscores. It is nonempty and does not start with an underscore. In othe...

Anonymous

3:03 AM

@ATaco It's Gibibytes. Capital B indicates bytes, lowercase b indicates bits.

Dennis

But it's 32 bits per integer, not 32 bytes.

ATaco

I'm questioning your math rather than your phrasing, 32*2**31-1 would be for a 32 bit integer, but it's 2**31-1 bytes, so should it not be 4*2**31-1?

Anonymous

@Dennis Oh crap, mixed units incorrectly

Anonymous

@ATaco You're right

ATaco

(Giving 2**33-1)

Dennis

3:08 AM

2**33-4 since we're at it.

ATaco

Oh right, because FOIL

Dennis

.oO(English math has a lot of crazy acronyms.)

ATaco

It's a really stupid way of saying "When two sets of terms are multiplied together, each term is multiplied against all of the other's terms"

ASCII-only

@ATaco Also a much more concise way

ATaco

Anyway, we can rather quickly prove that such an array is impossible. Because x86 (And this is probably still true for most 32bit architectures) has pointer sizes of 4 bytes, or 32 bits (Shocker), which means the last entry in memory can be no larger than 2^32-1 (actually this time, because this is unsigned), which is just a bit too short :P

Dennis

3:14 AM

x86 is perfectly capable of addressing more than 4 GiB of memory.

ATaco

Probably, but I have no idea how it works past there.

Dennis

Or do you really think 8-bit CPUs can only address 256 bytes of RAM?

ATaco

That's a bit condescending.

Dennis

Sorry, not intended that way.

ATaco

Neither CPUs ever really need to deal with plain arrays of size greater than their page size.

Dennis

3:17 AM

It's a common misconception since 32-bit Windows doesn't allow you to use more than 4 GiB of RAM. Thanks to PAE, that's an arbitrary limit, only designed to make you buy more expensive versions of Windows.

ATaco

I've only very recently been playing with Assemblers and languages that compile to Machine Code.

Dennis

The way you usually find out is when you buy more RAM but the OS says it isn't there. :P

Or used to, anyway, since 32-bit desktop computer are now a thing of the past.

ATaco

And If I'm correct, my 64bit processor can handle 18446744 GiB of memory.

Which I don't see being a problem for a bit.

Dennis

Again, the actual amount of memory a 64-bit processor can handle is quite unrelated to 2**64.

ASCII-only

@ATaco You mean 16TiB?

Dennis

3:21 AM

Especially once you factor in the OS.

ATaco

but whyy!

C says my pointers are 8 bytes, so 64bit, and that's the limit of the thinking I plan to do.

ASCII-only

Your pointers don't have to be 64 bits...

Dennis

For example, 64-bit Windows 10 Home can use up to 128 GiB of memory. Windows 10 Pro goes up to 2 TiB. This has nothing to do with pointer size.

ATaco

Oh, OS limitations would effect that I suppose.

Effect? Affect?

TuxCopter

affect

ASCII-only

3:24 AM

64-bit computing

In computer architecture, 64-bit computing is the use of processors that have datapath widths, integer size, and memory address widths of 64 bits (eight octets). Also, 64-bit computer architectures for central processing units (CPUs) and arithmetic logic units (ALUs) are those that are based on processor registers, address buses, or data buses of that size. From the software perspective, 64-bit computing means the use of code with 64-bit virtual memory addresses. However, not all 64-bit instruction sets support full 64-bit virtual memory addresses; x86-64 and ARMv8, for example, support only 48...

Affect

Downgoat

@ATaco it would be latter

Dennis

Physical Address Extension

In computing, Physical Address Extension (PAE), sometimes referred to as Page Address Extension, is a memory management feature for the x86 architecture. PAE was first introduced by Intel in the Pentium Pro, and later by AMD in the Athlon processor. It defines a page table hierarchy of three levels (instead of two), with table entries of 64 bits each instead of 32, allowing these CPUs to directly access a physical address space larger than 4 gigabytes (232 bytes). The page table structure used by x86-64 CPUs when operating in long mode further extends the page table hierarchy to four levels...

Downgoat

@Dennis these are limits Microsoft has imposed though, no?

Dennis

@Downgoat Yes, precisely. The 4 GiB limit that is grounds for the common n-bit pointers allow to address exactly 2^n bytes of memory misconception is another example. 32-bit Windows Server 2003 Enterprise gets you 64 GiB of RAM.

So you can break the magic 4 GiB mark, assuming that it's worth a lot of money to you.

Or get a sane OS without these silly restrictions.

9

ATaco

I don't want to think about how memory management works at that level

Dennis

3:33 AM

Fair enough. The important thing to remember is that 32-bit register doesn't equal 4 GiB of RAM. Nevermind the details.

ASCII-only

3:52 AM

@ASCII-only RIP turns out my hard drive is fine and all, it's still the connections that are broken

Maltysen

@Dennis does it just use 2 registers to store each pointer?

ASCII-only

@Maltysen No. Pls read the Wikipedia article Dennis linked above ty.

Maltysen

ok

Anonymous

In a flat memory addressing scheme, you can only address 2**(pointer_size) bytes of memory. However, basically no OS nowadays actually uses a flat memory addressing scheme.

Maltysen

@Mego I still don't get it

how can an n bit pointer refer to more than 2**n things

Dennis

4:06 AM

It can't, but the NULL pointer itself can point anywhere.

Maltysen

hmm?

Dennis

Think of pointers as house numbers. Different houses in different streets can have the same number.

Maltysen

@Dennis yeah, i get that, msdn is saying that PAE ads a third layer of indirection

ATaco

0xBAADF00D is my favourite pointer.

Maltysen

but how can that information, the 'street number' and 'house number' all be contained in 32 bits

Anonymous

4:10 AM

@Maltysen In a flat addressing scheme without virtual memory, each physical byte is given an address. With virtual memory, the same thing happens, but a two-tier page table is used to translate virtual addresses to physical addresses. PAE uses a 3-tier page table, which allows an extra level of indirection, so that each process can have its own 4 GB chunk, which may or may not overlap with another process's chunk.

Dennis

@Maltysen They aren't. The application still sees only 4 GiB. Which 4 GiB is handled by the OS.

Maltysen

@Mego oh, so each process still only can access 4gb virtual memoery

that makes sense

thanks

ATaco

There are probably ways to see more than one page of memory, but I have no clue how.

Dennis

Using multiple processes.

Anonymous

@Maltysen Yep. And with multiple processes and MPC, you can have more than 4 GiB at once.

Anonymous

4:14 AM

x86-64 with PAE and native long mode goes even crazier - a fourth level is added to the page table, allowing 64 GB of virtual memory for each process, and up to 256 TB total addressable memory.

ATaco

Good guess g++

Maltysen

@Mego aren't pointers usually 64 bits in x86-64?

or were u referring to the PAE making things crazy

Anonymous

@Maltysen Both. Wider pointers = more addresses, extra table in PAE = more addressable memory supported

Maltysen

k

Anonymous

So with x86-64 with native long mode, you can have 4096 processes, each using 64 GiB of virtual memory

Anonymous

4:18 AM

(or more processes using less memory each)

Dennis

Which should be enough for everybody.

Anonymous

@Dennis Famous last words

Anonymous

I'm sure Google can find a way to make Chrome need more than 64 GiB :P

Dennis

By Chrome 100, it will probably cache the entire internet.

ASCII-only

@Mego *each tab

ATaco

4:21 AM

> 640K ought to be enough for anybody.

New Main Posts

0

Q: heroX SocialCausalityPrizeI

The objective of the SociaCausalityPrizeI, now open at the X-Prize Foundation's heroX site, is to create the smallest program that outputs the CSV file of the USA Counties Data File Downloads provided by CenStats of the US Census. The smallest such program is, by algorithmic information theory, ...

kolmogorov-complexity

Anonymous

@NewMainPosts Smells like spam

ATaco

Reading through it, it's not spam, it's just really poorly formatted.

TuxCopter

@NewMainPosts are they asking to print an 87MiB file?!?

ATaco

(I jumped the gun however, and already flagged it)

Pavel

4:40 AM

@TuxCopter tio.run/##K07MLi1K/P/f4/9/AA

TuxCopter

@Pavel :o

Anonymous

@ATaco I'm not convinced it isn't. It's an unwanted advertisement for a foundation's donation page.

Pavel

> 2 lines

TuxCopter

It seems to count UTF8 bytes, so there's probably a newline somewhere in the codepoints

ASCII-only

> 1 sloc

Pavel

4:53 AM

Well, yeah, that does appear to be one significant line of code.

Where's the insignificant line though.

Anonymous

@Pavel Trailing newline?

Pavel

@Mego probably, but:

in talk.tryitonline.net, Jun 10 at 6:53, by Dennis

@Phoenix Because as far as *nix is concerned, the linefeed is the last character of the line, not a character between two lines. https://tio.run/##S0oszvj/vyQxWUFZoSi1LLWoOFUhJzMvtfj//8SkZK6U1DSu9IxMAA

Anonymous

5:14 AM

@Pavel Then I have no clue

ATaco

5:46 AM

Good work coke.

Pavel

<nobr> is depracated

Pavel

6:00 AM

@Mego 20 seconds ago, I recieved a push notifaction for this reply, and it woke me up. Goddammit StackExchange app.

Well, I wasn't really asleep that fast, but it was still annoying.

New Main Posts

0

Q: How puzzle and code golf works

Hello, I have spent looking at the stackexchange sites and especially this has seemed quite interesting. But I would like you to explain to me how it works and what it is about, I know how to program in many languages, but in the answers I always see very rare syntaxes. I also see that they are...

code-golf

ASCII-only

6:14 AM

@Pavel The SE app is really behind, you should ask for more frequent (or at lease user-configurable) polling :P

New Meta Posts

0

Q: How puzzle and code golf works

Hello, I have spent looking at the stackexchange sites and especially this has seemed quite interesting. But I would like you to explain to me how it works and what it is about, I know how to program in many languages, but in the answers I always see very rare syntaxes. I also see that they are...

discussion code-golf

New Main Posts

7:07 AM

-2

Q: fuzzy number close to A*B

Suppose I am given two fuzzy numbers Fuzzy number close to 2 = 0.5/0 + 0.67/1 + 1/2 + 0.67/3 and Fuzzy number close to 5 = 1/3 + 0.75/4 + 1/5 + 0.25/6 How do I get Fuzzy number close to 10

math

Neil

7:51 AM

@Dennis I'd read that many consumer device drivers aren't tested under PAE so that only the server OSes have it enabled (because server device manufacturers would have done more testing).

@Pavel WSUS 3.0.2 does that too. (Windows 10 really mucked up WSUS in a big way...)

New Sandboxed Posts

9:17 AM

0

A: Sandbox for Proposed Challenges

The Best Language to Golf In? Your task is to search all PPCG Questions for ones that match the format: Contain the code-golf tag Have at least one accepted answer Using this info above create a leaderboard of the languages used across PPCG and score them by the number of times they are the ...

New Main Posts

0

Q: Undress a string

We already now how to strip a string from its spaces. However, as proper gentlemen, we should rather undress it. Undressing a string is the same as stripping it, only more delicate. Instead of removing all leading and trailing spaces at once, we remove them one by one. We also alternate betwe...

code-golf string

miles

9:32 AM

codegolf.stackexchange.com/a/143377/6710 and so it begins...

New Sandboxed Posts

9:50 AM

0

A: Sandbox for Proposed Challenges

Implement a BrainFlump interpreter BrainFlump is the latest alternate memory model brainfuck-esque turing tarpit. It operates on a memory model we call a "Dump", which is simply an un-ordered collection of integers, with a pointer indicating the current item to operate on. As it is "unordered",...

Mayube

@cairdcoinheringaahing See ^ for new spec of BrainFlump, I'll make a new pypy interpreter at some point

Mayube

10:15 AM

ugh, sed is such a bitch when it comes to inserting blocks of code into a script

ASCII-only

@Mayube This is what vim is for

Mayube

Yes, let me spend hours using vim to insert a block of code into a file on ~400 servers

Mayube

10:33 AM

to give some context: We have around 240 customers, most of which have 2 servers. The Lead Dev considers git too risky to use on live servers, and we don't use any other sort of deployment system, meaning when it comes time to roll out an update, the list of all 240 systems is split between the 4 of us, and we spend several hours sshing onto each system and running a script on them.

And this isn't some simple update-the-system script, it's a custom script that manually modifies or replaces each file as needed, which means for files that might have bespoke changes in per customer, we have to write sed statements that insert the changes without overwriting the entire file

also I'm the only dev on the team who actually uses the git repos we have set up for development

I'm honestly starting to despise this job just out of the complete lack of organization or structure within the dev process

ASCII-only

How is git too risky >_>

Plus wouldn't patches work perfectly for that purpose @Mayube

Mayube

because this one time the dev accidentally did a git checkout -b and lost all his changes

ASCII-only

>_> git doesn't let you do that unless you force it to

plus there's git citool

Mayube

@ASCII-only yes it does

ASCII-only

Hmm maybe, but it most certainly doesn't throw changes

Neil

10:41 AM

how does checkout -b lose changes?

ASCII-only

Plus git is complicated, if you have people like that then use Mercurial

Mayube

if you make a bunch of changes, then do git checkout -b myNewBranch it creates branch myNewBranch from master, and any uncommitted changes are lost

ASCII-only

@Mayube Changes are most definitely not lost

Neil

um... no

ASCII-only

Git would never do that

Neil

10:42 AM

they remain uncommitted changes

Mayube

@ASCII-only uncommitted changes are

ASCII-only

@Mayube As I said, git wouldn't let you unintentionally lose uncommitted changes

Neil

(I agree about its complexity though, I keep thinking that the work I do would be much easier in Mercurial)

ASCII-only

Almost everyone agrees about its complexity :P

Mayube

there's only one other dev on the team who knows enough about git to be able to use it without fucking everything up, so the others just keep blindly sharing fabricated horror stories about "this one time" they accidentally deleted the master branch

Neil

10:45 AM

in fact, for one project that I use, I have so many local changes, that when I want to update, I stash the changes, pull from git, commit to mercurial, update to previous commit in mercurial, reset in git, unstash changess, update in mercurial, reset in git

ASCII-only

haha

Geobits

@Mayube If you decide to leave (for that reason), make sure your new job prospect is actually better at it, rather than just saying they follow the practices. Interviewers lie (or sometimes have no idea), and at least half of the shops I've seen are terrible at stuff like this.

Neil

... because this is way easier than trying to do it in git alone

Mayube

@Geobits of the 4 dev companies I've worked for so far, this is the only one that has had such a poor sense of structure and organization

Geobits

You've been lucky :P

Mayube

10:50 AM

Don't get me wrong 2 of the other companies were shit too, but for other reasons

Geobits

Haha, ok fair enough

ASCII-only

@Mayube Why are you working for such bad companies >_>

Mayube

one was basically microsoft tech support, and the other was run by a salesman who had a habit of MASSIVELY underestimating dev time, and as a result under-quoting customers, and then expecting us to be able to build a completely bespoke eCommerce website in 8 hours because "that's what we quoted"

Geobits

Hey, whatever gets the sale, yeah? >_>

Mayube

@ASCII-only it's a vicious cycle. I can't get a job at a decent company because my resume makes it look like I don't want to work anywhere longer than 6 months

ASCII-only

10:53 AM

@Mayube Fix your resume >_>

Mayube

that would require working somewhere longer than 6 months

Geobits

<_<

ASCII-only

Right

@Mayube ~~Then try to find work at a promising startup~~ Well RIP I guess, I have no idea about what it's like in the UK

Geobits

Or, you know, straight up lying. It's a resume. Its only function is to get you in the door, and will in most cases never be looked at again.