> Big companies will often tithe to these megachurches. Some churches are bigger than others. The Linux Foundation makes hundreds of millions of dollars. Smaller foundations like the Python Software Foundation have to make do with only a few million.
This hides essential detail that would seem to very much weaken the argument. You have the Linux Foundation and the Mozilla Foundation that "make hundreds of millions of dollars", and then everyone else is orders of magnitude smaller. Python might be in third place, for all I know (or maybe it's Apache).
> It shows how most open source projects aren’t some giant megachurch like group. These projects are one person.
> It’s easy to assume everyone else is also a megachurch member, even if they are not. The church members are pretty noisy and get a lot of attention.
I suspect most of those random bazaar vendors would like to have a respectable church-sized building. Or at least a proper stall.
> If you look at modern day open source, it sometimes feels like the megachurch open source is better because they have a nice parking lot, give out donation receipts, and it doesn’t smell like kabobs.
Well, no; it has more to do with the sense that outsiders are taking the bazaar seriously.
The ASF, chartered as a 501(c)(3) nonprofit charity which serves the public good, has a budget a fraction the size of those of orgs chartered as 501(c)(6) nonprofits which serve the common business interests of members.
A quick check implies Apache is on the order of half the size, though. When I wrote the other comment it was just the only other name that came to mind.
> those random bazaar vendors would like to have a respectable church-sized building.
I believe the analogy breaks down here some. That is, actual bazaar vendors may want this (I suppose), but FOSS maintainers may or may not want an organization to form around them. They may be content with the way things are; or they may just want a co-maintainer.
I think most of them want some measure of success and notoriety. I'd imagine the large majority never even get a PR from a stranger. Long tail, you know.
It was a bad essay at the time and I don't think you can make a good essay by trying to build off it. Adding "megachurch" to the already strained metaphor didn't improve it.
As you point out in your linked comment, the original essay captured the zeitgeist of the time. It also influenced and inspired many people. From that perspective, it's hard for me to agree that it was bad. However, I don't think the content was original at the time (perhaps that's what you mean by bad?) - in the sense that ESR wasn't out ahead of people blazing some new trail and it also didn't hold up very well factually.
Yeah, it's worth remembering that at the time a compiler cost $10k+, an OS $1000s/year - you couldn't work on OS or compiler work unless you worked for a big hardware company - a whole lot of interesting work was locked away from most programmers
Wasn’t Cathedral and the Bazaar originally published in 1999? Who was paying thousands of dollars a year for an OS in 199? And I think GCC was already widespread by then, no?
I didn’t start programming until a few years later, but for sure by 2002, it seemed to me a given that compilers were free. It was my impression that stuff like Borland was niche and that serious stuff like Java and C were free.
Not saying you are wrong, just your comment surprised me. Maybe I have a revisionist memory or maybe those intervening 3 years were quite transformational in the industry.
I recall stuff like the Intel icc compiler being expensive and desirable, and things like client access licenses, hardware licenses (to allow using non-trivial amounts of RAM and multi-processing) and support plans for proprietary OSes being rather expensive. Consulting a SCO Unix price sheet from that era, a license that allowed 150 users and up to 32GB of RAM was $10k.
Prices also varied around OS features used. Vendors loved to nickel-and-dime you (separate *-user client licenses for file services, print services, remote access, etc), generally to drive you towards bigger packages that seemed like a better deal.
The firm I was at in 1997 was shipping commercial software with GCC. There were expensive compilers, but you weren't required to use them. For Windows builds, I think we were Borland C++, which was hundreds of dollars. Sun had a pretty expensive compiler for Solaris that I remember using for hunting down memory leaks.
Yes, that is the context in which I first read it (likely around 1999 when it appeared on slashdot), as a senior in high school with no access to the tools used by most professional programmers at the time.
Yes, I'm speaking about my experience as I remember it - not what was objectively possible for someone with the right resources and knowledge at the time :)
Right, I'm not so much pushing back on you as I am establishing a chronology for CATB. Ordinary people were absolutely belting out (what we would now call) open source software by the time it was written.
(That's not the biggest flaw in the essay, of course. It made predictions, some of which turned out to be comically wrong. The true parts of it weren't new, and the new parts of it weren't true.)
I guess it depends on what you think the goal of the essay was. I always felt like the primary goal was to inspire people and a lot of the software engineering parts were more framing. To me it reads as a manifesto disguised as a software engineering essay.
If you take the goal as inspiring people, i think it achieved its goals and then some. I'm pretty sure that CATB brought more people into FOSS than the GNU manifesto ever did.
(1) That rule refers to things people have posted to HN in things like "Show HNs" (or their moral equivalents). It isn't a general prohibition on critique, which would be silly.
(2) You may have missed the link to ~1,000 words of detailed criticism of CATB, on which I support my claim here that CATB is bad.
> (1) That rule refers to things people have posted to HN in things like "Show HNs" (or their moral equivalents).
There’s nothing I’m seeing in the text as it is written that suggests this to be the case. There are just a lot of comments I see that amount to: “I don’t like this,” which can be an interesting signal by itself but not if users refuse to elaborate on it, which is what I (erroneously) thought was happening here.
> You may have missed the link to ~1,000 words of detailed criticism of CATB, on which I support my claim here that CATB is bad.
I did miss it, sorry. I clicked through and didn’t notice that the top comment was yours. I assumed you were just linking to a past discussion.
I’m sure you already know this, but on the off chance you don’t, you can click on a comment’s timestamp to get a permalink to the specific comment, like this:
HN is a common law system; the real guidelines are the guidelines page itself, and the "jurisprudence" of years and years of Dan (and Tom) writing moderator comments. But you also know you're a little off the rails when you've derived a rule that would prohibit, say, criticism of a book --- "Teach Yourself C In 24 Hours is a bad book". Of course that's OK!
But yeah, the big thing here is that the substance of my critique is on a different thread. It's disfavored to retype things you can just link to. I'd be irritated with me too if I just said "CATB is bad!" and left it at that.
People are still talking about a flat Earth and creationism. Given 8 billion people, there are enough available braincells to keep even the stupidest idea floating around in the memesphere.
So the case that you're making here is that CATB is renowned amongst the kind of practitioners who think NULL pointers are "just how the hardware works". Sounds about right.
I know you're replying to a brand new (likely troll) account, but I'm also very confused by this and would be curious to learn if there's any truth to it. I personally don't really see what a Von Neumann machine has to do with null pointers (or how an implication would go either way), but maybe I'm missing something.
NULL pointers working the way they do was a design decision made my hardware engineers a long time ago because it saved some transistors when that mattered. We’re past that point now for most ASICs and hardware can be changed. Although backward software compatibility is a thing too.
Null pointers have nothing to do with the instruction set architecture, except as far as they are often represented by the value 0. Can you describe the scheme you're imagining, whereby their use saves transistors?
No, the CPU doesn't have a special pointer value which is designated invalid (except as far as modern address spaces are so large that you cannot possibly map memory to each address without mirroring). In many OSs, e.g. CP/M, address 0 is actually meaningful. The C idiom of cramming sum-type semantics into the nooks and crannies of a return value that ordinarily means something entirely different is an extremely poor one, and null pointers are the poster child: Tony Hoare's billion-dollar mistake.
It's absolutely fine to have a packed representation of a sum type "under the hood": this is how Rust implements Option<&T> (where T: Thin), for example. It's also fine to expose the layout of this packed representation to the programmer, as C's union does. But it's a huge footgun to have unchecked casts as the default. If not for this terrible convention, C wouldn't have any unchecked implicit casts: something like f(1 + 0.5) performs a coercion, a far more sensible behaviour.
The only reason we're talking about null pointers at all is because they were an influential idea, not because they were a good idea. Likewise with the essay.
While it's narrowly true that CPU instruction sets generally don't have a null-pointer concept, I'm not sure how important that is: the null pointer seems to have been (I don't know enough to be sure) a well-established idiom in assembly programming which carried across naturally to BCPL and C. (In much the same way that record types were, apparently, a common assembly idiom long before they became particularly normal to have in HLLs.) Programmers like being able to null out a pointer field, 0 is an obvious "joker" value, and jump-if-0 instructions tend to be convenient and fast. Whether or not you'd want to say it's "how the hardware works" it does seem to have a certain character of inevitability. Even if the Bell Research guys had disapproved of the idiom they would likely have had difficulty keeping it out of other people's C programs once C became popular. The Hoare ALGOL W thing seems to be more relevant to null pointers in Java and the like.
> Programmers like being able to null out a pointer field, 0 is an obvious "joker" value, and jump-if-0 instructions tend to be convenient and fast.
And there's nothing wrong with that! But you should write it
union {
char *ptr;
size_t scalar;
} my_nullable_pointer;
if (my_nullable_pointer.scalar) {
printf("%s", my_nullable_pointer.ptr);
}
not:
char *my_nullable_pointer;
if (my_nullable_pointer) {
printf("%s", my_nullable_pointer);
}
Yes, this takes up more space, but it also makes the meaning of the code clearer. typedef in a header can bring this down to four extra lines per pointer type in the entire program. Add a macro, and it's five extra lines plus one extra line per pointer type. Put this in the standard library, and the programmer has to type a few extra characters – in exchange for it becoming extremely obvious (to an experienced programmer, or a quick-and-dirty linter) when someone's introduced a null pointer dereference, and when a flawed design makes null pointer dereferences inevitable.
> The Hoare ALGOL W thing seems to be more relevant to null pointers in Java and the like.
I believe you are correct; but I like blaming Tony Hoare for things. He keeps scooping me: I come up with something cool, and then Tony Hoare goes and takes credit for it 50 years in the past. Who does he think he is, Euler?
> No, the CPU doesn't have a special pointer value which is designated invalid
Sort of right, sort of wrong.
From my understanding: older, simpler, architectures treat memory location zero as a normal memory address. On x86 and x64, the OS can configure the MMU to treat certain pages as invalid. Many years ago, I ran across a reference to Sparcs treating accesses to memory location zero as invalid. In other words, it depends upon which architecture you're dealing with.
Skimming appendix H of https://courses.grainger.illinois.edu/cs423/sp2011/lectures/..., I can't see any special treatment of the zero page, but https://stackoverflow.com/a/22847758/5223757 contains an anecdote about SPARCs not placing a page of zeroes at that address. I expect that's probably an OS restriction, and they considered it safer to modify the in-house software they understood, rather than tinker with the externally-sourced OS's memory management routines, but the anecdote is weak evidence that it might have been a hardware distinction at one point.
On the otherhand, I greatly appreciate that we don't pretend everyone is 100% awesome all the time. We shouldn't hold people up as role models that we don't want to emulate, and whatnot.
One of them is legit a saint and the other almost as much. They absolutely are role models, and the way they are talked about now is exactly a lesson in the problem. If more people emulated them, the world would be a much better place.
If you're talking about Eric S. Raymond here, I'm having trouble not believing that this is just bait. Even in the Linux community, purely on Linux terms he's a problematic and polarizing figure.
I think enough of us have imperfections that we can appreciate that people who've done wonderful things have also done some very $#!tty things. Someone doesn't need to be a saint to still have a wide, positive influence.
I absolutely think we should put people in bad boxes, and would go so far as to suggest if you are worried about this possibility, you may be worried about which box you belong in. ;)
I think the important part is there must always remain a possibility for someone to exit that box. Repentance and forgiveness are key values in themselves, and we must be able to accept people if they can change.
Unrepentant garbage people who still make garbage statements and do garbage things, however, can remain in the dumpster where they belong until such time they warrant climbing out.
> you may be worried about which box you belong in. ;)
There’s also the risk someone very loud decides to put you in a box you don’t belong in. Eventually you are able to demonstrate it, but, in the meantime, you need to deal with the consequences.
Your post may be insinuating that you put ESR and RMS in such boxes, although you did not actually say that. You might want to clarify that point. (And I say that as someone who has neither upvoted or downvoted you.)
I'll also say that there are enough aspects of our personality and behavior that you might use to justify placing someone in the "bad box" that almost everyone would be in one; and if you were to relax the criteria so that you "average badness" along multiple axes, that comes with its own problems.
I always interpreted cathedral vs bazaar as being about the architecture of large things. Do you build to a master plan? Or does everyone do whatever they want? (Within some kind of framework, of course.) Like the cathedral of the Java SDKs vs the flea market of NPM.
This author seems to have some kind of attitude about organization in general—anything with people and process, that happens to exist around some project, that might require at least a small commitment to be a part of. Like complaining that a flea market has a form to sign.
The ability for people to functionally collaborate, with some kind of structure, is the key thing that enables building large things together.
Yes, and well done as well. Unlike the other two unmentionables, Linus very much worthy of remembrance. Sure he was extra grumpy for a long time but that's about the only bad thing you can say about the man.
The post referred to the Sovereign Tech Agency (https://www.sovereign.tech). The problem that the Sovereign Tech Agency is trying to solve seems to be a hard one.
yet 7 days ago someone who works on OpenPrinting was here and stated:
"The whole printing stack is supported by 4 people, 2 of whom are doing that since the inception of CUPS in 1999. Scanning is maintained by a single person."
idk, without the sovereign tech agency it would be fewer people, or they would have less time to work on the project. You can't expect the German government to completely fill any need for resources in open source software.
This makes me wonder - is there some platform on which people who maintain important (or arguably-important) facilities can post Wanted ads for volunteer co-maintainers?
I realize that the number of people who would actually be crazy enough to browse that platform and answer such ads is pretty small... but - it may be noticeably above Zero.
I like the idea that we moved from cathedrals to megachurches because it explains why everything feels so corporate now. It is easy to forget that the messy bazaar is still underneath all the shiny tools we use.
Large endeavours require some level of “megachurchness”. Linux back then was tiny in comparison with what it is today. So was Python. Nowadays we have much larger projects that encompass a much larger space than we had in the 1990s. You can’t make things consistent at these sizes without some governance in place.
There are still a lot of space for projects without much structure- if you have NSA codenames that aren’t public yet (and you are not subject to US laws) you can contribute with the nsaname tool and have cool names for your servers and containers. If you want to help adding glyphs to my 3278 font, you can. You can do that to millions of small projects that are small enough to not require much structure.
The Cathedral metaphor doesn't make any sense since the point of the Cathedral is simultaneously to revere God and to be able to take in as many "unwashed masses" as possible. Only by self-exclusion (explicit external irreverence/scandal) can you be excluded.
It works for me. Cathedral is analogous to free software being a religion. It is a theocratic worldview that has a zealous following that must apply the rituals of old. Bazaar is the marketplace. It is supposed to be a efficient market metaphor for software being transactional and not relational.
Is this a perfect metaphor? I think its a rigid way of looking at software on either side. I think it is more grey. I like the merits of both sides.
That is not what Eric S. Raymond (esr) was describing.
GNUnix was developed using the Cathedral-style, Linux was developed using the bazaar-style. How Linux development was coordinated was thought to be impossible for something that had to be as solid as an operating system. The essay is a deep dive, exploring the conditions that the Linux project needed to ship an OS.
But ESR believed in right wing, libertarian adjacent politics. He's advocating for deregulated, free market ideas in the form of criticizing GNU. In doing this, he was seeking out the preferred metaphor and working backwards, rather than describing what is.
The author links to another article of theirs called "Open Source is Bigger Than You Can Imagine," which hinges on the size of the npm registry. npm says "open source" on their landing page, and has an "npm Open Source" section of their policies, which places no restrictions on how you license your npm package (save for a special license to them).
This does seem very bazaar to me, but this would all be deemed Not Open Source by the [cathedral/megachurch?] community, correct? Do people take issue with npm using the term open source?
If we're working with those metaphors, I think it's useful to read up on how actual, real-life bazaars are operating.
In particular:
> A bazaar or souk is a marketplace consisting of multiple small stalls or shops [...] They are traditionally located in vaulted or covered streets that have doors on each end and served as a city's central marketplace.
> Merchants specialized in each trade were also organized into guilds, which provided support to merchants but also to clients. The exact details of the organizations varied from region to region. Each guild had rules that members were expected to follow, but they were loose enough to allow for competition. Guilds also fulfilled some functions similar to trade unions and were able to negotiate with the government on behalf of merchants or represent their interests when needed.
> Historically, in Islamic cities, the muḥtasib was the official in charge of regulating and policing the bazaar and other aspects of urban life. They monitored things such as weights and measures, pricing, cleanliness, noise, and traffic circulation, as well as being responsible for other issues of public morality. They also investigated complaints about cheating or the quality of goods.
So not quite the anarchocapitalist, self-organizing utopia that tech people seem to imagine there - in fact, they have a lot of organization, both between merchants as well as on the bazaar as a whole.
Seems to me, this model is more similar to the "privately-owned marketplaces" we see increasingly in the digital world: App stores, merchant sites like Amazon, etc.
In that sense, "most of open-source" being on Github which is now owned by Microsoft is ironically more similar to a real bazaar.
With one difference: At least the administrators of real bazaars were public officials with a mandate to keep the market fair - and there was organization among the vendors in form of guilds. With digital marketplaces, the markets themselves are private assets and the administrators are blatantly self-interested. And there doesn't seem to be any kind if higher-order organization across different open source projects, everyone is fighting on their own.
So maybe it would do the open source community good to become more like an actual bazaar.
>Seems to me, this model is more similar to the "privately-owned marketplaces" we see increasingly in the digital world: App stores, merchant sites like Amazon, etc.
>In that sense, "most of open-source" being on Github which is now owned by Microsoft is ironically more similar to a real bazaar.
Id put it that this is incorrect insofar - as the bazaar was/is a public commons with a dual regulatory environment city(state) and the guilds , which would enforce/regulate as needed.
The digital marketplaces we have would be more anologous to feudal plantations ,where each coder(sharecropper) survives at the whim of their particluar feudal lord , who have total control within that space and the state via lobbying mostly keeps off.Theer are no guild equivalent so when Playstore/Github makes a ruling like the recent hike of dev fees or ci runner. Theres no state or user leverage that can force a reversal other than complaints.
Paradoxically id say they are more megachurch than bazaars.
Yep and its insane when most devs are actively hostile to unins etc from too much libertarian koolaid when they can see the active backing things like teacher/nurse/police unions provide.
They may have some bad ideas , butthe structure and backing kinda gets glossed over.
With that title, I'm clicking and reading all the way through.
I'm writing an article on a similar topic, but it's a critique on a popular development style that imports a huge dependency supply chain (without concern on if they are cathedral, bazaar, or megachurches), and what the benefits of building your thing bottom-up has.
If this sounds interesting to you, hacker news reader, you can leave a comment and I'll reply with a link once it's published.
The latest thing though is that the megachurches send out these evangelist priests who run an inquisition into your amounts tithed. These people then go around trying to co-opt the machinery of the state to redirect money to the megachurches.
“We should tax everyone to fund open source” they say
“Google should pay a percentage of their gross revenue to the Rust Software Foundation” they say
All this is because it’s enough for the bazaar to create but the author has correctly identified that the purpose of the megachurches is to receive tithes.
The Rust megachurch is one of the biggest proponents of this and its adherents are always trying to take our money by force because we won’t give it by will https://news.ycombinator.com/item?id=46048954
Free and open source software provide a ton of value to businesses and consumers. It's right that tax dollars is used to fund what effectively is a public good so that we can all benefit from it even more.
I can see a government requiring itself to provide some funding to open-source projects that it actually makes a lot of use of. But not just open-source in general; no one needs to get funding for some pet project that only that one person cares about and isn't very good anyway: putting some crappy chatGPT-generated code on GitHub should not qualify you for government funding.
There's always a cause and a church. There is an instrument for this: your donations can be tax deductible if you give to a 501c3 that exists for the public benefit. But that's not enough for you guys. Having seen the success of private equity dialysis clinics to redirect Medicare funding, you have decided that you want a piece of this government revenue pie. Enough of this greed.
Don't both, people like them hold society back. I suggest you go out and talk to your physical neighbors about taxing big tech, it has a huge amount of support. The only question is do you want a democratic administration to use said tax revenues to benefit the public or a republican administration to benefit a few private actors.
It's going to happen and I know what side I'd rather be on.
Other countries killed 250k/month in gas chambers till we came to liberate them, and they still exist under our shield. Their opinions are kind of moot.
> The TL;DR was that old open source was the cathedral of exclusive developers and groups. Then the Bazaar showed up (which was the Linux Kernel for example) and that freed us from the shackles of the cathedral.
I didn't make it past the tldr lol is this some kind of poisoned data for GPT 6?
There's a other group besides these: the secret society, who infiltrate the cathedrals, the megachurches and the bazaar. They are quite cultish, but thankfully the "Data Primacy Lodge" is gaining more initiates than the old guard "Order of Objects"
> "...Microsoft. Who we haven’t mentioned in this story, but they hated Linux more than a toddler hates naps."
A lot of FOSS people think this but it's not really true. It was a thorn in the side of MS executives as a competitor, sure, but I never met anyone in the rank and file that could be bothered to hate Linux. More than a few of my colleagues played with Linux at home in the '00s. I cut my teeth on the commercial UNIXes so there wasn't anything interesting about Linux to me until it had caught up with them around 2010 or so.
you're trying to rewrite history here, Microsoft used to be a well known linux hater, but linux became popular and they had no choice but to accept it. Remember the "linux is cancer" years...
Might want to schedule an appointment for a neurologist because acting like MSFT wasn't anti linux is revisionist history that borders on medical intervention.
I was there, too, and I remember all of the FUD from MS. I remember the Halloween documents, MS funding SCO’s lawsuit, etc. MS saw Linux as a threat, especially in the server space. The goal was to stomp it out, like they did to Netscape.
> Big companies will often tithe to these megachurches. Some churches are bigger than others. The Linux Foundation makes hundreds of millions of dollars. Smaller foundations like the Python Software Foundation have to make do with only a few million.
This hides essential detail that would seem to very much weaken the argument. You have the Linux Foundation and the Mozilla Foundation that "make hundreds of millions of dollars", and then everyone else is orders of magnitude smaller. Python might be in third place, for all I know (or maybe it's Apache).
> It shows how most open source projects aren’t some giant megachurch like group. These projects are one person.
> It’s easy to assume everyone else is also a megachurch member, even if they are not. The church members are pretty noisy and get a lot of attention.
I suspect most of those random bazaar vendors would like to have a respectable church-sized building. Or at least a proper stall.
> If you look at modern day open source, it sometimes feels like the megachurch open source is better because they have a nice parking lot, give out donation receipts, and it doesn’t smell like kabobs.
Well, no; it has more to do with the sense that outsiders are taking the bazaar seriously.
The ASF, chartered as a 501(c)(3) nonprofit charity which serves the public good, has a budget a fraction the size of those of orgs chartered as 501(c)(6) nonprofits which serve the common business interests of members.
The PSF is also 501(c)(3) (https://www.python.org/psf/mission/).
A quick check implies Apache is on the order of half the size, though. When I wrote the other comment it was just the only other name that came to mind.
> those random bazaar vendors would like to have a respectable church-sized building.
I believe the analogy breaks down here some. That is, actual bazaar vendors may want this (I suppose), but FOSS maintainers may or may not want an organization to form around them. They may be content with the way things are; or they may just want a co-maintainer.
I think most of them want some measure of success and notoriety. I'd imagine the large majority never even get a PR from a stranger. Long tail, you know.
It was a bad essay at the time and I don't think you can make a good essay by trying to build off it. Adding "megachurch" to the already strained metaphor didn't improve it.
https://news.ycombinator.com/item?id=35939383
As you point out in your linked comment, the original essay captured the zeitgeist of the time. It also influenced and inspired many people. From that perspective, it's hard for me to agree that it was bad. However, I don't think the content was original at the time (perhaps that's what you mean by bad?) - in the sense that ESR wasn't out ahead of people blazing some new trail and it also didn't hold up very well factually.
Yeah, it's worth remembering that at the time a compiler cost $10k+, an OS $1000s/year - you couldn't work on OS or compiler work unless you worked for a big hardware company - a whole lot of interesting work was locked away from most programmers
Wasn’t Cathedral and the Bazaar originally published in 1999? Who was paying thousands of dollars a year for an OS in 199? And I think GCC was already widespread by then, no?
I didn’t start programming until a few years later, but for sure by 2002, it seemed to me a given that compilers were free. It was my impression that stuff like Borland was niche and that serious stuff like Java and C were free.
Not saying you are wrong, just your comment surprised me. Maybe I have a revisionist memory or maybe those intervening 3 years were quite transformational in the industry.
I recall stuff like the Intel icc compiler being expensive and desirable, and things like client access licenses, hardware licenses (to allow using non-trivial amounts of RAM and multi-processing) and support plans for proprietary OSes being rather expensive. Consulting a SCO Unix price sheet from that era, a license that allowed 150 users and up to 32GB of RAM was $10k.
Prices also varied around OS features used. Vendors loved to nickel-and-dime you (separate *-user client licenses for file services, print services, remote access, etc), generally to drive you towards bigger packages that seemed like a better deal.
The firm I was at in 1997 was shipping commercial software with GCC. There were expensive compilers, but you weren't required to use them. For Windows builds, I think we were Borland C++, which was hundreds of dollars. Sun had a pretty expensive compiler for Solaris that I remember using for hunting down memory leaks.
Yes, that is the context in which I first read it (likely around 1999 when it appeared on slashdot), as a senior in high school with no access to the tools used by most professional programmers at the time.
FreeBSD 2.0 was 1994.
Yes, I'm speaking about my experience as I remember it - not what was objectively possible for someone with the right resources and knowledge at the time :)
Right, I'm not so much pushing back on you as I am establishing a chronology for CATB. Ordinary people were absolutely belting out (what we would now call) open source software by the time it was written.
(That's not the biggest flaw in the essay, of course. It made predictions, some of which turned out to be comically wrong. The true parts of it weren't new, and the new parts of it weren't true.)
It was certainly influential. It's just bad on its own merits.
I guess it depends on what you think the goal of the essay was. I always felt like the primary goal was to inspire people and a lot of the software engineering parts were more framing. To me it reads as a manifesto disguised as a software engineering essay.
If you take the goal as inspiring people, i think it achieved its goals and then some. I'm pretty sure that CATB brought more people into FOSS than the GNU manifesto ever did.
> Please don't post shallow dismissals, especially of other people's work.
https://news.ycombinator.com/newsguidelines.html
(1) That rule refers to things people have posted to HN in things like "Show HNs" (or their moral equivalents). It isn't a general prohibition on critique, which would be silly.
(2) You may have missed the link to ~1,000 words of detailed criticism of CATB, on which I support my claim here that CATB is bad.
> (1) That rule refers to things people have posted to HN in things like "Show HNs" (or their moral equivalents).
There’s nothing I’m seeing in the text as it is written that suggests this to be the case. There are just a lot of comments I see that amount to: “I don’t like this,” which can be an interesting signal by itself but not if users refuse to elaborate on it, which is what I (erroneously) thought was happening here.
> You may have missed the link to ~1,000 words of detailed criticism of CATB, on which I support my claim here that CATB is bad.
I did miss it, sorry. I clicked through and didn’t notice that the top comment was yours. I assumed you were just linking to a past discussion.
I’m sure you already know this, but on the off chance you don’t, you can click on a comment’s timestamp to get a permalink to the specific comment, like this:
https://news.ycombinator.com/item?id=35940773
HN is a common law system; the real guidelines are the guidelines page itself, and the "jurisprudence" of years and years of Dan (and Tom) writing moderator comments. But you also know you're a little off the rails when you've derived a rule that would prohibit, say, criticism of a book --- "Teach Yourself C In 24 Hours is a bad book". Of course that's OK!
But yeah, the big thing here is that the substance of my critique is on a different thread. It's disfavored to retype things you can just link to. I'd be irritated with me too if I just said "CATB is bad!" and left it at that.
You're completely wrong. The fact that people are still talking about it today proves it has some kind of worth. The essay was great.
People are still talking about a flat Earth and creationism. Given 8 billion people, there are enough available braincells to keep even the stupidest idea floating around in the memesphere.
People are still talking about null pointers: that doesn't mean they were ever a good idea.
That's just how the hardware works. Don't like it? Make your own CPU.
So the case that you're making here is that CATB is renowned amongst the kind of practitioners who think NULL pointers are "just how the hardware works". Sounds about right.
I know you're replying to a brand new (likely troll) account, but I'm also very confused by this and would be curious to learn if there's any truth to it. I personally don't really see what a Von Neumann machine has to do with null pointers (or how an implication would go either way), but maybe I'm missing something.
It has nothing to do with NULL pointers and is instead a property of a programming language.
NULL pointers working the way they do was a design decision made my hardware engineers a long time ago because it saved some transistors when that mattered. We’re past that point now for most ASICs and hardware can be changed. Although backward software compatibility is a thing too.
Null pointers have nothing to do with the instruction set architecture, except as far as they are often represented by the value 0. Can you describe the scheme you're imagining, whereby their use saves transistors?
No, the CPU doesn't have a special pointer value which is designated invalid (except as far as modern address spaces are so large that you cannot possibly map memory to each address without mirroring). In many OSs, e.g. CP/M, address 0 is actually meaningful. The C idiom of cramming sum-type semantics into the nooks and crannies of a return value that ordinarily means something entirely different is an extremely poor one, and null pointers are the poster child: Tony Hoare's billion-dollar mistake.
It's absolutely fine to have a packed representation of a sum type "under the hood": this is how Rust implements Option<&T> (where T: Thin), for example. It's also fine to expose the layout of this packed representation to the programmer, as C's union does. But it's a huge footgun to have unchecked casts as the default. If not for this terrible convention, C wouldn't have any unchecked implicit casts: something like f(1 + 0.5) performs a coercion, a far more sensible behaviour.
The only reason we're talking about null pointers at all is because they were an influential idea, not because they were a good idea. Likewise with the essay.
While it's narrowly true that CPU instruction sets generally don't have a null-pointer concept, I'm not sure how important that is: the null pointer seems to have been (I don't know enough to be sure) a well-established idiom in assembly programming which carried across naturally to BCPL and C. (In much the same way that record types were, apparently, a common assembly idiom long before they became particularly normal to have in HLLs.) Programmers like being able to null out a pointer field, 0 is an obvious "joker" value, and jump-if-0 instructions tend to be convenient and fast. Whether or not you'd want to say it's "how the hardware works" it does seem to have a certain character of inevitability. Even if the Bell Research guys had disapproved of the idiom they would likely have had difficulty keeping it out of other people's C programs once C became popular. The Hoare ALGOL W thing seems to be more relevant to null pointers in Java and the like.
> Programmers like being able to null out a pointer field, 0 is an obvious "joker" value, and jump-if-0 instructions tend to be convenient and fast.
And there's nothing wrong with that! But you should write it
not: Yes, this takes up more space, but it also makes the meaning of the code clearer. typedef in a header can bring this down to four extra lines per pointer type in the entire program. Add a macro, and it's five extra lines plus one extra line per pointer type. Put this in the standard library, and the programmer has to type a few extra characters – in exchange for it becoming extremely obvious (to an experienced programmer, or a quick-and-dirty linter) when someone's introduced a null pointer dereference, and when a flawed design makes null pointer dereferences inevitable.> The Hoare ALGOL W thing seems to be more relevant to null pointers in Java and the like.
I believe you are correct; but I like blaming Tony Hoare for things. He keeps scooping me: I come up with something cool, and then Tony Hoare goes and takes credit for it 50 years in the past. Who does he think he is, Euler?
> No, the CPU doesn't have a special pointer value which is designated invalid
Sort of right, sort of wrong.
From my understanding: older, simpler, architectures treat memory location zero as a normal memory address. On x86 and x64, the OS can configure the MMU to treat certain pages as invalid. Many years ago, I ran across a reference to Sparcs treating accesses to memory location zero as invalid. In other words, it depends upon which architecture you're dealing with.
Skimming appendix H of https://courses.grainger.illinois.edu/cs423/sp2011/lectures/..., I can't see any special treatment of the zero page, but https://stackoverflow.com/a/22847758/5223757 contains an anecdote about SPARCs not placing a page of zeroes at that address. I expect that's probably an OS restriction, and they considered it safer to modify the in-house software they understood, rather than tinker with the externally-sourced OS's memory management routines, but the anecdote is weak evidence that it might have been a hardware distinction at one point.
They aren't there in asm.
There are lots of proven bad ideas still being bandies about today, and it does not prove they are anything but enduringly worthless.
"Don't look him up, he's not exactly role model material." I don't admire the ethos of putting people in bad boxes.
On the otherhand, I greatly appreciate that we don't pretend everyone is 100% awesome all the time. We shouldn't hold people up as role models that we don't want to emulate, and whatnot.
If we're not pretending everyone is awesome then why permanently deselect certain people as role models?
One of them is legit a saint and the other almost as much. They absolutely are role models, and the way they are talked about now is exactly a lesson in the problem. If more people emulated them, the world would be a much better place.
If you're talking about Eric S. Raymond here, I'm having trouble not believing that this is just bait. Even in the Linux community, purely on Linux terms he's a problematic and polarizing figure.
You believe in the commentary and advocacy he provides for pedophilia? Why should I and others pay more attention to that?
There isn't any such "advocacy".
I think enough of us have imperfections that we can appreciate that people who've done wonderful things have also done some very $#!tty things. Someone doesn't need to be a saint to still have a wide, positive influence.
Which wonderful things are you referring to?
I absolutely think we should put people in bad boxes, and would go so far as to suggest if you are worried about this possibility, you may be worried about which box you belong in. ;)
I think the important part is there must always remain a possibility for someone to exit that box. Repentance and forgiveness are key values in themselves, and we must be able to accept people if they can change.
Unrepentant garbage people who still make garbage statements and do garbage things, however, can remain in the dumpster where they belong until such time they warrant climbing out.
> you may be worried about which box you belong in. ;)
There’s also the risk someone very loud decides to put you in a box you don’t belong in. Eventually you are able to demonstrate it, but, in the meantime, you need to deal with the consequences.
Your post may be insinuating that you put ESR and RMS in such boxes, although you did not actually say that. You might want to clarify that point. (And I say that as someone who has neither upvoted or downvoted you.)
I'll also say that there are enough aspects of our personality and behavior that you might use to justify placing someone in the "bad box" that almost everyone would be in one; and if you were to relax the criteria so that you "average badness" along multiple axes, that comes with its own problems.
I always interpreted cathedral vs bazaar as being about the architecture of large things. Do you build to a master plan? Or does everyone do whatever they want? (Within some kind of framework, of course.) Like the cathedral of the Java SDKs vs the flea market of NPM.
This author seems to have some kind of attitude about organization in general—anything with people and process, that happens to exist around some project, that might require at least a small commitment to be a part of. Like complaining that a flea market has a form to sign.
The ability for people to functionally collaborate, with some kind of structure, is the key thing that enables building large things together.
> History will probably remember him as LTT, “Linus The Torvalds”
This is trolling right?
There are a lot of tangential, one-liner, throwaway jokes in this article.
Yes, everyone knows LTT is Linus Tech Tips!
> This is trolling right?
Yes, and well done as well. Unlike the other two unmentionables, Linus very much worthy of remembrance. Sure he was extra grumpy for a long time but that's about the only bad thing you can say about the man.
The post referred to the Sovereign Tech Agency (https://www.sovereign.tech). The problem that the Sovereign Tech Agency is trying to solve seems to be a hard one.
OpenPrinting is listed as a funded project:
https://www.sovereign.tech/tech/openprinting
yet 7 days ago someone who works on OpenPrinting was here and stated:
"The whole printing stack is supported by 4 people, 2 of whom are doing that since the inception of CUPS in 1999. Scanning is maintained by a single person."
https://news.ycombinator.com/item?id=46579361
Isn't this the situation the Sovereign Tech Agency is trying to avoid?
idk, without the sovereign tech agency it would be fewer people, or they would have less time to work on the project. You can't expect the German government to completely fill any need for resources in open source software.
Yikes :-(
This makes me wonder - is there some platform on which people who maintain important (or arguably-important) facilities can post Wanted ads for volunteer co-maintainers?
I realize that the number of people who would actually be crazy enough to browse that platform and answer such ads is pretty small... but - it may be noticeably above Zero.
I like the idea that we moved from cathedrals to megachurches because it explains why everything feels so corporate now. It is easy to forget that the messy bazaar is still underneath all the shiny tools we use.
It's a great phrase which explains how a company like Apple can take good ideas, implement the parts they like, and not give back.
Large endeavours require some level of “megachurchness”. Linux back then was tiny in comparison with what it is today. So was Python. Nowadays we have much larger projects that encompass a much larger space than we had in the 1990s. You can’t make things consistent at these sizes without some governance in place.
There are still a lot of space for projects without much structure- if you have NSA codenames that aren’t public yet (and you are not subject to US laws) you can contribute with the nsaname tool and have cool names for your servers and containers. If you want to help adding glyphs to my 3278 font, you can. You can do that to millions of small projects that are small enough to not require much structure.
The Cathedral metaphor doesn't make any sense since the point of the Cathedral is simultaneously to revere God and to be able to take in as many "unwashed masses" as possible. Only by self-exclusion (explicit external irreverence/scandal) can you be excluded.
The metaphor does not refer to the finished building but to the building process
It works for me. Cathedral is analogous to free software being a religion. It is a theocratic worldview that has a zealous following that must apply the rituals of old. Bazaar is the marketplace. It is supposed to be a efficient market metaphor for software being transactional and not relational.
Is this a perfect metaphor? I think its a rigid way of looking at software on either side. I think it is more grey. I like the merits of both sides.
That is not what Eric S. Raymond (esr) was describing.
GNUnix was developed using the Cathedral-style, Linux was developed using the bazaar-style. How Linux development was coordinated was thought to be impossible for something that had to be as solid as an operating system. The essay is a deep dive, exploring the conditions that the Linux project needed to ship an OS.
But ESR believed in right wing, libertarian adjacent politics. He's advocating for deregulated, free market ideas in the form of criticizing GNU. In doing this, he was seeking out the preferred metaphor and working backwards, rather than describing what is.
The author links to another article of theirs called "Open Source is Bigger Than You Can Imagine," which hinges on the size of the npm registry. npm says "open source" on their landing page, and has an "npm Open Source" section of their policies, which places no restrictions on how you license your npm package (save for a special license to them).
This does seem very bazaar to me, but this would all be deemed Not Open Source by the [cathedral/megachurch?] community, correct? Do people take issue with npm using the term open source?
The article says "GNU's not Linux". No, it's "GNU's not Unix".
It's so confidently written too, lol. Like. Think about what an acronym is and if it could possibly stand for something with an L...
You can't correct humour.
When something is obviously wrong, perhaps learn to ask yourself if it's trying to be funny. Is dead Python funny?
If we're working with those metaphors, I think it's useful to read up on how actual, real-life bazaars are operating.
In particular:
> A bazaar or souk is a marketplace consisting of multiple small stalls or shops [...] They are traditionally located in vaulted or covered streets that have doors on each end and served as a city's central marketplace.
> Merchants specialized in each trade were also organized into guilds, which provided support to merchants but also to clients. The exact details of the organizations varied from region to region. Each guild had rules that members were expected to follow, but they were loose enough to allow for competition. Guilds also fulfilled some functions similar to trade unions and were able to negotiate with the government on behalf of merchants or represent their interests when needed.
> Historically, in Islamic cities, the muḥtasib was the official in charge of regulating and policing the bazaar and other aspects of urban life. They monitored things such as weights and measures, pricing, cleanliness, noise, and traffic circulation, as well as being responsible for other issues of public morality. They also investigated complaints about cheating or the quality of goods.
( https://en.wikipedia.org/wiki/Bazaar )
So not quite the anarchocapitalist, self-organizing utopia that tech people seem to imagine there - in fact, they have a lot of organization, both between merchants as well as on the bazaar as a whole.
Seems to me, this model is more similar to the "privately-owned marketplaces" we see increasingly in the digital world: App stores, merchant sites like Amazon, etc.
In that sense, "most of open-source" being on Github which is now owned by Microsoft is ironically more similar to a real bazaar.
With one difference: At least the administrators of real bazaars were public officials with a mandate to keep the market fair - and there was organization among the vendors in form of guilds. With digital marketplaces, the markets themselves are private assets and the administrators are blatantly self-interested. And there doesn't seem to be any kind if higher-order organization across different open source projects, everyone is fighting on their own.
So maybe it would do the open source community good to become more like an actual bazaar.
>Seems to me, this model is more similar to the "privately-owned marketplaces" we see increasingly in the digital world: App stores, merchant sites like Amazon, etc.
>In that sense, "most of open-source" being on Github which is now owned by Microsoft is ironically more similar to a real bazaar.
Id put it that this is incorrect insofar - as the bazaar was/is a public commons with a dual regulatory environment city(state) and the guilds , which would enforce/regulate as needed.
The digital marketplaces we have would be more anologous to feudal plantations ,where each coder(sharecropper) survives at the whim of their particluar feudal lord , who have total control within that space and the state via lobbying mostly keeps off.Theer are no guild equivalent so when Playstore/Github makes a ruling like the recent hike of dev fees or ci runner. Theres no state or user leverage that can force a reversal other than complaints.
Paradoxically id say they are more megachurch than bazaars.
Guilds are now scorned as communism
Yep and its insane when most devs are actively hostile to unins etc from too much libertarian koolaid when they can see the active backing things like teacher/nurse/police unions provide. They may have some bad ideas , butthe structure and backing kinda gets glossed over.
With that title, I'm clicking and reading all the way through.
I'm writing an article on a similar topic, but it's a critique on a popular development style that imports a huge dependency supply chain (without concern on if they are cathedral, bazaar, or megachurches), and what the benefits of building your thing bottom-up has.
If this sounds interesting to you, hacker news reader, you can leave a comment and I'll reply with a link once it's published.
The latest thing though is that the megachurches send out these evangelist priests who run an inquisition into your amounts tithed. These people then go around trying to co-opt the machinery of the state to redirect money to the megachurches.
“We should tax everyone to fund open source” they say
“Google should pay a percentage of their gross revenue to the Rust Software Foundation” they say
All this is because it’s enough for the bazaar to create but the author has correctly identified that the purpose of the megachurches is to receive tithes.
The Rust megachurch is one of the biggest proponents of this and its adherents are always trying to take our money by force because we won’t give it by will https://news.ycombinator.com/item?id=46048954
Rust delenda est.
Free and open source software provide a ton of value to businesses and consumers. It's right that tax dollars is used to fund what effectively is a public good so that we can all benefit from it even more.
I can see a government requiring itself to provide some funding to open-source projects that it actually makes a lot of use of. But not just open-source in general; no one needs to get funding for some pet project that only that one person cares about and isn't very good anyway: putting some crappy chatGPT-generated code on GitHub should not qualify you for government funding.
There's always a cause and a church. There is an instrument for this: your donations can be tax deductible if you give to a 501c3 that exists for the public benefit. But that's not enough for you guys. Having seen the success of private equity dialysis clinics to redirect Medicare funding, you have decided that you want a piece of this government revenue pie. Enough of this greed.
Rust delenda est.
Alright Cato, but consider that other countries successfully spend their budgets on public goods like infrastructure and the arts.
Don't both, people like them hold society back. I suggest you go out and talk to your physical neighbors about taxing big tech, it has a huge amount of support. The only question is do you want a democratic administration to use said tax revenues to benefit the public or a republican administration to benefit a few private actors.
It's going to happen and I know what side I'd rather be on.
Other countries killed 250k/month in gas chambers till we came to liberate them, and they still exist under our shield. Their opinions are kind of moot.
Rust delenda est.
This is inane. There are other countries besides Germany that finance public works.
Normally I'd say read a history book, but it might be quicker to read a newspaper.
Doctor, heal thyself.
Rust delenda est.
Kind of offtopic but fun fact I didn't know until recently, the Moldbug definition of Cathedral is based (lol) on the Eric Raymond definition
I stalled on Which is an acronym for “Gnu’s not Linux” and can't recover from the spin.
It's actually "Gnu's Not Unix", the original article got it wrong too
the article also says the creator of Linux is LTT
It's a joke
It's just harmless trolling
Given the tone of the article I’m sure it was tongue-in-cheek humor and not an error.
It's a joke. I think.
It's excellent. I grinned ear to ear.
> The TL;DR was that old open source was the cathedral of exclusive developers and groups. Then the Bazaar showed up (which was the Linux Kernel for example) and that freed us from the shackles of the cathedral.
I didn't make it past the tldr lol is this some kind of poisoned data for GPT 6?
Not sure if GPT played a role, but for one the editor did a poor job. Very sloppy writing indeed
There's a other group besides these: the secret society, who infiltrate the cathedrals, the megachurches and the bazaar. They are quite cultish, but thankfully the "Data Primacy Lodge" is gaining more initiates than the old guard "Order of Objects"
> "...Microsoft. Who we haven’t mentioned in this story, but they hated Linux more than a toddler hates naps."
A lot of FOSS people think this but it's not really true. It was a thorn in the side of MS executives as a competitor, sure, but I never met anyone in the rank and file that could be bothered to hate Linux. More than a few of my colleagues played with Linux at home in the '00s. I cut my teeth on the commercial UNIXes so there wasn't anything interesting about Linux to me until it had caught up with them around 2010 or so.
People mean Microsoft, the corporation, as a policy. Not every employee there literally.
you're trying to rewrite history here, Microsoft used to be a well known linux hater, but linux became popular and they had no choice but to accept it. Remember the "linux is cancer" years...
I was there a couple decades and you weren't.
The devs weren't, but
https://www.theregister.com/2001/06/02/ballmer_linux_is_a_ca...
Microsoft messaging was very clear at the time
Might want to schedule an appointment for a neurologist because acting like MSFT wasn't anti linux is revisionist history that borders on medical intervention.
I was there, too, and I remember all of the FUD from MS. I remember the Halloween documents, MS funding SCO’s lawsuit, etc. MS saw Linux as a threat, especially in the server space. The goal was to stomp it out, like they did to Netscape.