Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Please don't use Slack for FOSS projects

drewdevault.com

Post reply on HN
Print view View original

221–230 of 499 posts

Re: Please don't use Slack for FOSS projects

#221

Post by TazeTSchnitzel »

upofadown wrote: post #192
vjeux wrote: post #163

Earlier quoted context omitted.

IRC usage stats are really bad. Steady decline from 1 million users in 2003 to 400k in 2012. In a time where we're talking about communication apps with hundreds of millions of users, not even having a million is pretty telling to me. http://royal.pingdom.com/2012/04/24/irc-is-dead-long-live-ir...

A lot of IRC usage used to be purely social. So the fact that that are lots less users isn't really indicative of anything in the context of FOSS development.

Indeed, in the same time period I think Freenode actually grew.

Re: Please don't use Slack for FOSS projects

#223

Post by jsprogrammer »

>That the reasoning is fallacious doesn't make the conclusion incorrect.

This is trivially true by the [material implication, modus ponens, modus tollens, other name, etc] truth table. However, if you study the truth table a bit longer, you will realize that you should stop reading any text which contains fallacious or self-contradictory logic. While the conclusion may be "correct", it is correct despite all of the other text you read (so, go find a logical argument [instead of relying on known bad information]). The text you read can tell you nothing about whether the conclusion is true.

Fallacious reasoning needs to be identified, targeted, and destroyed.

Edit: I've been rate-limited, so any fallacious (or non) arguments posted in response will need to wait a while for a proper response (some have already been typed up).

Re: Please don't use Slack for FOSS projects

#224

Post by brazzledazzle »

daenney wrote: post #143
brazzledazzle wrote: post #139

One thing even private organizations should think about is the potential for chats on Slack to embarrass the heck out of them. Slack doesn't let an organization sunset and delete private chats/DMs after a certain amount of time. When employees think they aren't overheard they start to get very frank with each other (which is good, I think frank and direct communication is important). But sometimes they say things tha…

Looking at this page it seems entirely possible to configure a DM retention policy: https://slack.zendesk.com/hc/en-us/articles/203457187-Settin... It seems to by default use the team retention settings.

Thanks a bunch for this. I'm surprised the person helping me didn't know about this (unless it's fairly new) but I should have put more effort into researching it myself.

Re: Please don't use Slack for FOSS projects

#225

Post by KirinDave »

mbillie1 wrote: post #207
KirinDave wrote: post #130

If IRC were good enough to handle the needs of small software company communication, people would use IRC . Sitting here pretending everyone just doesn't know about a 20+ year old technology is comical. The "IRC Features" section bugs me, too. Unless you run a private IRC server IRC has a long, tortured history of design-based security difficulties. Gone are the days of netsplits permanently owning channels, but the…

> If IRC were good enough to handle the needs of small software company communication, people would use IRC. Sitting here pretending everyone just doesn't know about a 20+ year old technology is comical. I'd just like to point out that this argument ("if were so great, everyone would already use/do ") is just a slightly different form of the ancient appeal to popularity argument and is entirely fallacious.

I didn't mean to imply it was causal. I implied that the industry is full of people who know about the product comparison and make exactly the opposite value judgement.

Whatever gave you the impression I think this is causal?

Re: Please don't use Slack for FOSS projects

#227

Post by KirinDave »

jsprogrammer wrote: post #223

>That the reasoning is fallacious doesn't make the conclusion incorrect. This is trivially true by the [material implication, modus ponens, modus tollens, other name, etc] truth table. However, if you study the truth table a bit longer, you will realize that you should stop reading any text which contains fallacious or self-contradictory logic. While the conclusion may be "correct", it is correct despite all of the o…

Indeed. So if someone doesn't actually make a claim, but then another party begins to rebut said claim as if it were part of a formal argument, that's called a Straw Man Argument or "Burning a straw man" colloquially.

That's what you're doing. We should identify it, target it, and destroy it.

Actually I don't really think that it should be destroyed. It's human nature. We find any post-facto justifications for our gut reactions all the time. If we can't find one, we tend to distort what we saw to create one. It's part of human cognition.

Re: Please don't use Slack for FOSS projects

#228

Post by netcan »

jedberg wrote: post #160

This is the perfect counter example to "Open source is better". IRC has been around for 20+ years. But a company that is just a few years old blows it away feature-wise. Sure, you can do all the same stuff in IRC, but look right at the article for why you wouldn't. The answer to every "missing feature" in IRC is to install some extra software and then maintain it. I'm don't want to waste time doing that, I've got pro…

I generally agree, though I'm not sure financial incentive is the reason. Open source has really shown that financial incentive is (A) complex and (B) not the only thing that motivates people.

If you look at open and closed source across a lot of categories, I think it's more common to see OS "win" where the product is more objective. When it come to more subjective problems like UX, it seems to be more difficult.

If your goal is something fairly objective OS is really amazing. Take VLC's "play everything everywhere" goal. Fantastic, objective goal for and OS project that really plays to its strengths.

Linux is, I guess, the canonical example. It's fantastically successful relative to closed source competitors on every front that is objective. But, its consistently been unsuccessful as a consumer desktop.

That said, I don't think IRC is dead either. Some people/communities prefer IRC.

Re: Please don't use Slack for FOSS projects

#229

Post by KirinDave »

simias wrote: post #154
KirinDave wrote: post #130

If IRC were good enough to handle the needs of small software company communication, people would use IRC . Sitting here pretending everyone just doesn't know about a 20+ year old technology is comical. The "IRC Features" section bugs me, too. Unless you run a private IRC server IRC has a long, tortured history of design-based security difficulties. Gone are the days of netsplits permanently owning channels, but the…

IRC is used by many projects small and big and I've used it in a few companies as well, it's not like it's an absurd concept. I don't know if Slack is better or worse (I've never used it, I thought the title was talking about Slackware linux which seemed odd) but you draw a much bleaker portrait of IRC than what I experience day to day using it extensively. Now if you want to talk about mailing lists I might side wit…

> but you draw a much bleaker portrait of IRC

I'm being brutally real. IRC is a brittle, aged, poorly designed protocol even by the standards of 1988 academia. You could design a more robust centralized service just by using off the shelf components today.

Seriously. Use off the shelf Google Container Engine stuff along with some modest websocket work and even relative neophyte will end up with a more robust service with better scaling characteristics.

Really, that's why stuff like Slack and Hipchat exist. Becuase it's not really all that hard to do a much better implementation than what IRC can feasibly do.

Post reply on HN