That's a wonderful combination of self-aware and clueless. Heaps praise about how wonderful and smart his coworkers are, then says "I'm sure they'd tell me not to publish this, but I'm gonna...
p.s. My partners and coworkers at Cognitect were not consulted regarding this message - I am certain they would have dissuaded me. These opinions are mine alone.
That's a wonderful combination of self-aware and clueless.
Heaps praise about how wonderful and smart his coworkers are, then says "I'm sure they'd tell me not to publish this, but I'm gonna disregard that because I think it's bad advice".
I actually agree with this. Putting an OSS-licensed public repo on Github in no way implies putting up with the public. If a project is officially community run, that's great. I have seen it work....
I actually agree with this. Putting an OSS-licensed public repo on Github in no way implies putting up with the public. If a project is officially community run, that's great. I have seen it work. But nobody is required to do that, and looks like the Clojure folks chose not to. If you don't like a leadership decision, then either make something better yourself or fork the repo and maintain your own tree.
I'm honestly quite surprised by many of the reactions here on Tildes. I thought the article was a good reminder that OSS is simply a licensing mechanism and that anything on top of that is a...
I'm honestly quite surprised by many of the reactions here on Tildes. I thought the article was a good reminder that OSS is simply a licensing mechanism and that anything on top of that is a bonus.
Maybe the reactions differ based on whether or not you've ever maintained a popular OSS project? I've got a couple libraries I maintain and this resonated strongly with me. The level of entitlement some (not all, but some) people have when using free software is incredible.
Some very good food for thought on FOSS Governance is provided by Mozilla: https://blog.mozilla.org/blog/2018/05/15/whats-your-open-source-strategy-here-are-10-answers/ esp. read the linked PDF....
Some very good food for thought on FOSS Governance is provided by Mozilla:
The feeling of entitlement is definitely an issue. But simply dumping stuff on the ’net and slapping a license on it is not governing your FOSS either (see “Bathwater” Open Source Archetype).
Good article, I'll have to read it in-depth later. That said, I don't think Hickey is saying that he wants to run Clojure using the "Bathwater" model (clearly it is run much better than that atm)....
Good article, I'll have to read it in-depth later.
That said, I don't think Hickey is saying that he wants to run Clojure using the "Bathwater" model (clearly it is run much better than that atm). I read it more that he's taking aim at people who feel entitled to determine Clojure's governance archetype, which is definitely overstepping your bounds IMO - it's up to the maintainers to determine governance.
The major failure with this memo is delivery. He's rude, arrogant and offensive with this. Another problem is that they were not up-front about this. I thought Clojure was proper open-source...
The major failure with this memo is delivery. He's rude, arrogant and offensive with this.
Another problem is that they were not up-front about this. I thought Clojure was proper open-source software, but they're doing it more like Google or Microsoft. That's fine, but that's apparently not how everybody perceived it. See the tweet I quoted in another comment under this thread, for example. That is from a person heavily invested in the Clojure community, and one of the 3 authors of the O'Reilly Clojure book.
Clojure is proper open-source software as are many things from Microsoft and Google. What do you mean when you say "proper open-source software"? I think that's the crux of the issue.
I thought Clojure was proper open-source software
Clojure is proper open-source software as are many things from Microsoft and Google.
What do you mean when you say "proper open-source software"? I think that's the crux of the issue.
I think I underspecified there. I meant I thought it was more community oriented than it was (not necessarily community driven, though). The source of that misunderstanding is not merely the term...
I think I underspecified there. I meant I thought it was more community oriented than it was (not necessarily community driven, though). The source of that misunderstanding is not merely the term "open source" though, I have also followed the language for quite some time from its public inauguration. The way Hickey went from one conference to another, the content of the videos I watched of him, and the attitude of the community suggested a different picture of Clojure back then. Frankly, I only recently discover this sort of hostility towards the community. But I never really invested much time into the language, TBH.
I think that Clojure is pretty community oriented in a lot of ways. If you think of the project like a bus, they let the community on the bus, and try to make the ride on the bus palatable, but...
I think that Clojure is pretty community oriented in a lot of ways. If you think of the project like a bus, they let the community on the bus, and try to make the ride on the bus palatable, but they don't let the community drive the bus. The bus has a specific route that is planned out, and the driver knows what that route is, and even if the community says, "Hey, we need this bus to go somewhere totally different!" that probably won't happen.
First, I'm one of many maintainers of a very large open source project. This colours my opinion on entitlement. Sure, but it becomes a problem when people expect public governance, when none is...
First, I'm one of many maintainers of a very large open source project. This colours my opinion on entitlement.
But simply dumping stuff on the ’net and slapping a license on it is not governing your FOSS either
Sure, but it becomes a problem when people expect public governance, when none is implied. "Just send PRs", "just open an issue" and so on are good indicators of a project where anyone can contribute anything.
But sometimes, I just want to put stuff on Github in case someone somewhere finds it useful. That should be ok without facing the wrath of entitled masses or being required to entertain their ideas.
It's unfamiliar to me as well, but from some spelunking it seems like it's centered around this mailing list announcement spawning criticism, for example this Twitter thread. Reading between the...
Reading between the lines, it sounds like there's a core group of Clojure devs employed by Cognitect who are coming to consensus with each other on language design choices, but not necessarily consensus with the broader Clojure community. This post seems to be trying to spin that "we'd like to be involved in design decisions too" sentiment as people who are ungrateful for the hard work the devs are putting in.
"Users are stupid and I don't need them. I'm a demiurge for the plebes, provider of divine creations that the ignorant masses just leech off of my Apollonian body. They're entitled to nothing. I hate dealing with them, and that's why I publicly released my project on the largest software repository on the internet. But please, use my project. The stars look good on my resumé."
I mean, we aren't entitled to good software. We aren't entitled to working software, even. Since I'm not entitled to an explanation, either, I might as well guess on my own whether Clojure is good or working software, no? It seems to have a lot of people posting some one-line piece of idolizing praise unrelated to the content of the post. Again, with no explanation given, I could guess by the character of posts here that Clojure is some kind of mid-level marketing gambit or a cult. But if I was permitted explanations, perhaps I wouldn't think that.
Open source may be a gift, but it doesn't come with batteries included, and so there's a cost for adopting such a "gift." If you want people to use your software, closed- or open-source, you do need to make it worth their time and show some basic decency to your users. I do wonder what necessitated this bizarre post in the first place, but I suppose I'm not entitled to an answer. And since open source isn't a community, only a licensing and delivery mechanism, I'm not entitled to show you any adulation or even respect; you need to deserve it as an individual. You had some respect before from me for recognizing important concepts regarding immutability and identity and building a language that elegantly handles them, but this post isn't helping in the respect department.
Clojure is a circlejerk. A huge one. It's a half-baked dialect of Lisp, which can't have proper error handling, let alone a condition-restart system or proper recursion, and yet people go out of...
Clojure is a circlejerk. A huge one. It's a half-baked dialect of Lisp, which can't have proper error handling, let alone a condition-restart system or proper recursion, and yet people go out of their way to claim that Hickey is a genius. Look at the comments, look at the relevant HN thread, it's a huge circlejerk.
If you want to write Lisp on the JVM, there is Kawa Scheme and Armed Bear Common Lisp. Both can interoperate with Java etc., and both should have better errors than Clojure. Also, they are compliant to two major standard Lisp dialects as their names imply. All they lack is a privileged whiny bastard that has a little army of fanboys.
That's... harsh, to say the least. I use Clojure on a daily basis because I find that it's an effective tool to do the job. It's niche, sure, but used widely enough to have a solid toolchain and a...
That's... harsh, to say the least.
I use Clojure on a daily basis because I find that it's an effective tool to do the job. It's niche, sure, but used widely enough to have a solid toolchain and a decent array of libraries. There's enough traction that it's reasonable to expect answers on StackOverflow! I haven't found another Lisp that I can say the same for yet.
I do not refuse that, but it is not as featureful as Common Lisp or as clean and well defined as Scheme. There are certain places it is lacking. What I don't understand is why is it perceived as...
I do not refuse that, but it is not as featureful as Common Lisp or as clean and well defined as Scheme. There are certain places it is lacking. What I don't understand is why is it perceived as genial and/or revolutionary. It's a mediocre language that is mostly hype-driven.
Especially considering the timing of this with other events, this is so incredibly whiny.
See: https://github.com/dominictarr/event-stream/issues/116
That's a wonderful combination of self-aware and clueless.
Heaps praise about how wonderful and smart his coworkers are, then says "I'm sure they'd tell me not to publish this, but I'm gonna disregard that because I think it's bad advice".
I actually agree with this. Putting an OSS-licensed public repo on Github in no way implies putting up with the public. If a project is officially community run, that's great. I have seen it work. But nobody is required to do that, and looks like the Clojure folks chose not to. If you don't like a leadership decision, then either make something better yourself or fork the repo and maintain your own tree.
I'm honestly quite surprised by many of the reactions here on Tildes. I thought the article was a good reminder that OSS is simply a licensing mechanism and that anything on top of that is a bonus.
Maybe the reactions differ based on whether or not you've ever maintained a popular OSS project? I've got a couple libraries I maintain and this resonated strongly with me. The level of entitlement some (not all, but some) people have when using free software is incredible.
Some very good food for thought on FOSS Governance is provided by Mozilla:
https://blog.mozilla.org/blog/2018/05/15/whats-your-open-source-strategy-here-are-10-answers/
esp. read the linked PDF.
TL;DR main take-away for this thread:
The feeling of entitlement is definitely an issue. But simply dumping stuff on the ’net and slapping a license on it is not governing your FOSS either (see “Bathwater” Open Source Archetype).
Good article, I'll have to read it in-depth later.
That said, I don't think Hickey is saying that he wants to run Clojure using the "Bathwater" model (clearly it is run much better than that atm). I read it more that he's taking aim at people who feel entitled to determine Clojure's governance archetype, which is definitely overstepping your bounds IMO - it's up to the maintainers to determine governance.
The major failure with this memo is delivery. He's rude, arrogant and offensive with this.
Another problem is that they were not up-front about this. I thought Clojure was proper open-source software, but they're doing it more like Google or Microsoft. That's fine, but that's apparently not how everybody perceived it. See the tweet I quoted in another comment under this thread, for example. That is from a person heavily invested in the Clojure community, and one of the 3 authors of the O'Reilly Clojure book.
Clojure is proper open-source software as are many things from Microsoft and Google.
What do you mean when you say "proper open-source software"? I think that's the crux of the issue.
I think I underspecified there. I meant I thought it was more community oriented than it was (not necessarily community driven, though). The source of that misunderstanding is not merely the term "open source" though, I have also followed the language for quite some time from its public inauguration. The way Hickey went from one conference to another, the content of the videos I watched of him, and the attitude of the community suggested a different picture of Clojure back then. Frankly, I only recently discover this sort of hostility towards the community. But I never really invested much time into the language, TBH.
I think that Clojure is pretty community oriented in a lot of ways. If you think of the project like a bus, they let the community on the bus, and try to make the ride on the bus palatable, but they don't let the community drive the bus. The bus has a specific route that is planned out, and the driver knows what that route is, and even if the community says, "Hey, we need this bus to go somewhere totally different!" that probably won't happen.
First, I'm one of many maintainers of a very large open source project. This colours my opinion on entitlement.
Sure, but it becomes a problem when people expect public governance, when none is implied. "Just send PRs", "just open an issue" and so on are good indicators of a project where anyone can contribute anything.
But sometimes, I just want to put stuff on Github in case someone somewhere finds it useful. That should be ok without facing the wrath of entitled masses or being required to entertain their ideas.
I'm afraid I'm not up to speed on what's happening with Clojure. What's the context to this article?
It's unfamiliar to me as well, but from some spelunking it seems like it's centered around this mailing list announcement spawning criticism, for example this Twitter thread.
Reading between the lines, it sounds like there's a core group of Clojure devs employed by Cognitect who are coming to consensus with each other on language design choices, but not necessarily consensus with the broader Clojure community. This post seems to be trying to spin that "we'd like to be involved in design decisions too" sentiment as people who are ungrateful for the hard work the devs are putting in.
I really liked this tweet on it: https://twitter.com/cemerick/status/1067111260611850240
reply from the user tommyettinger:
Clojure is a circlejerk. A huge one. It's a half-baked dialect of Lisp, which can't have proper error handling, let alone a condition-restart system or proper recursion, and yet people go out of their way to claim that Hickey is a genius. Look at the comments, look at the relevant HN thread, it's a huge circlejerk.
If you want to write Lisp on the JVM, there is Kawa Scheme and Armed Bear Common Lisp. Both can interoperate with Java etc., and both should have better errors than Clojure. Also, they are compliant to two major standard Lisp dialects as their names imply. All they lack is a privileged whiny bastard that has a little army of fanboys.
That's... harsh, to say the least.
I use Clojure on a daily basis because I find that it's an effective tool to do the job. It's niche, sure, but used widely enough to have a solid toolchain and a decent array of libraries. There's enough traction that it's reasonable to expect answers on StackOverflow! I haven't found another Lisp that I can say the same for yet.
I do not refuse that, but it is not as featureful as Common Lisp or as clean and well defined as Scheme. There are certain places it is lacking. What I don't understand is why is it perceived as genial and/or revolutionary. It's a mediocre language that is mostly hype-driven.