There are plenty of problems with BDFL-style projects, but I think there are a lot of advantages too. Redis is unique in my experience in that it works the way you'd expect - it doesn't cause outages, it is fast, and it has a vanishingly small number of gotchas. The feature set is well curated and for the most part fits together cohesively.
The most important thing Antirez did, in my opinion, was to say "No" to things. No to new features that didn't make sense. No to PRs that didn't fit the vision. Saying no is one of the most important jobs of a project maintainer. It's a thankless task that upsets a lot of people. But it's critical for a project to stay successful and achieve the leader's vision.
Maybe I'm a pessimist, but I predict after a few years of this new model we'll see weird features, more stability issues, and performance regressions as more cooks enter the kitchen. Time will tell.
Redis has had tremendous mission creep over the years. It started of course mostly as a volatile cache but I've now seen it also used as a message bus (in three different ways: BLPOP, PUB/SUB and Streams), for service discovery and as a general purpose database - something that Redis Labs (in my opinion: wrongly) encourages.
Memcache existed in 2009 when Redis was first released, also as a volatile cache...and still is just a volatile cache. Memcache is what "saying no" looks like. Redis is what "saying yes" looks like - and there are a lot of gotchas.
Redis has stayed the same, architecturally, from the beginning: it’s a keyspace where the values are arbitrary in-memory objects owned by their keys, and where commands resolve to synchronous function-calls against the objects (or ref-cells) at those keys.
Anything that can be done without changing that architecture, is in-scope for Redis. (Though if a data structure doesn’t have wide use outside of a domain, it’s best left to a module.) Anything that cannot be done without changing the architecture, won’t be done.
Much of the “fun” I’ve personally had in watching Redis evolve, has been seeing how Antirez has managed to solve the puzzles of getting features that intuitively wouldn’t be a good fit for this architecture, to fit into it anyway: changing technologies that are implemented one way everywhere else, into something else for Redis, that still work, are still performant, and still solve isomorphic use-cases—if not with the same steps you’d use to solve them in other systems. (E.g. using Streams vs. using a regular MQ; using Redis Cluster vs. using regular RDBMS replication; etc.)
I personally have not enjoyed all the strange and wonderful ways people have found to use Redis. Most of them are pretty fragile and (most dangerously) they often put all Redis uses in the same instance...with eviction turned on.
Maybe that's what Cal and his colleagues were using it for at the time, but the claim that it started as a volatile cache is just not true. Interesting datastructures and the fork based persistence model were there from the start.
Lists, hashes, sets and such that you have available in your programming language, but available as a networked server so any language/app/process can access the same data using the same useful structures.
But... things that are also just useful often and for what you'd need to spin off a separate service for, or make suboptimally in whatever (No)SQL you use to store your data.
So it is perfect if you just want to prototype whether given approach is viable, and "good enough" for many apps.
It works excellent as a message bus, though. And you can add HSET/HDEL to your list as the fourth way.
Lots of reasons why but here is just one because I'm short on time: all of these (four) ways of doing a message bus with Redis offer either a work queue mode or publish/subscribe - but not both (perhaps excepting streams which I've forgotten some of the details of). I have yet to see many real world messaging scenarios where you don't end up using both.
A "proper" message bus needs to have topics, exchanges and queues - like Rabbit. 50% of that functionality is usually not 50% of the value but 10% or less (especially when you consider all the other things about Redis you need to think about - like how eviction is configured).
This is a bit typical of Redis: for any X, you can normally do X in Redis, but is it really a good idea to do X in Redis? IMO, usually no.
Redis Streams actually does exactly what you describe and is basically a single-node fast Kafka replacement with multiple independent consumer groups, per-message acknowledgements, and queue semantics. We use it successfully in many high-load scenarios.
BDFL governance is extremely underrated in the opensource world. People attribute the success of the Linux kernel to its open contribution style, but I will argue that Linux is successful because there's a dictator at the top that enforces a direction, a long term goal, and most importantly, says NO.
Community based governance is a direct cause of that jwz's Cascade of Attention-Deficit problem one can find all over the open source world, especially Linux desktop related.
We don't have enough data points to determine if what you are saying is correct. I have also seen lots of BDFL projects smoke out.
Is the lesson that actual dictatorships only survive long term if 1) the dictator is a nice person and 2) they eventually get managed by an open democratic process?
I think you are actually describing the two clocks problem. Lots of BDFL projects have their own Cascade of Attention-Deficit issues, the project only goes in the direction the BDFL actually cares about.
There is no Steve Jobs in a democratic committee.
> To make a user-friendly system, the system must have conceptual integrity, which can only be achieved by separating architecture from implementation. A single chief architect (or a small number of architects), acting on the user's behalf, decides what goes in the system and what stays out. The architect or team of architects should develop an idea of what the system should do and make sure that this vision is understood by the rest of the team. A novel idea by someone may not be included if it does not fit seamlessly with the overall system design. In fact, to ensure a user-friendly system, a system may deliberately provide fewer features than it is capable of. The point being, if a system is too complicated to use, many features will go unused because no one has time to learn them.
* (This is quoted from the Wikipedia page as I don't have an online copy of MMM to hand).
Heck at one of Postgres conferences there was a shirt that said something like "Tom Lane reviewed my patch and all I got was this stupid shirt."
While he was not a BDFL style person, in the past, if he didn't like a change, it didn't make it in.
Keeping both of those balls in the air at once is why being a software coordinator like this is such an exhausting job.
The structure of a committee project doesn't make it better or worse than BDFL. It just makes it more average. Any greatness present is toned down. And substandardness is brought up.
This also main reason the "led by greatness" BDFL projects are better than the committee run ones and the "led by less than great" BDFL projects "smoke out". The committee's tend towards the middle of the bell curve, BDFL's are all over the place.
No, but it doesn't have as much success as Linux either
Democracy is just the least worse political system.
IRL you own a plot of land, or a home. Or you have a small factory, and there are a lot of network dependencies, who does you sell the supplies, where you sell your products. It the local BDFL get's mad it is very difficult to move. (Specially if there are some cash flow controls were you can't get out with the money that you get selling your home or that you have saved.)
Let's say you have a fix for some driver used on ARM. You'll submit it via Phabricator to get ARM maintainers to review it. If they don't, because eg they don't have time, then you'll probably commit it anyway, but if someone points out you did something wrong, you'll be expected to fix it.
Imagine an author has gotten started writing volumes 1 and 2, and those volumes became popular — maybe the Harry Potter books as an example. Now, adding more authors with equal "story-line decision making power", or a committee, for volumes 3, 4, 5, is a weird idea, right.
Linux is made because of many people all over the world, but it is successful because of the BDFL at the top. A consistent sense of mission absolutely benefits a project in the long term. There just isn't anyone that's going to get your idea better than you, when you're doing something that you know enough about to seriously get started on.
In essence, Antirez has protected the core of Redis from needing to bloat, by making it possible for anyone who wants to build on top of Redis to literally do so—build on top, rather than inside. As such, I expect future PRs to Redis Core to look much like current PRs to Redis Core: just fixing bugs, adding stability, and resolving infrastructure-level interoperability concerns.
As for the future of Redis I choose to be more optimistic. I think the Redis philosophy (thanks to Antirez) has become ingrained. The benefits of a minimal feature set are hopefully well-established.
In the very least I suspect this will last much longer than many expect before the bureaucrats take over.
(This is also referenced and linked in the article.)
This is hard. It makes you feel bad inside to tell people "No", especially when they open PR and clearly did a lot of work adding some feature.
Reminds me of Brian Goetz: "We have to say no to almost anything. If we didn't, things would pretty quickly degenerate to the point where the system collapses on its own weight (...) if we did this, which of these eighteen possible features might we foreclose on, and might we want to leave those possibilities open, at the cost of saying no to this thing here." (https://www.infoq.com/presentations/java-history-present-fut..., 24m+)
Redis has succeeded because of antirez. I really wish it can continue doing so for the long run without him. More than anything it reminds me of Steve Jobs' death (although gladly this is in much happier circumstances and I wish antirez many more decades of hacking). It's true that a decade later Apple is still doing great, but you just know things would have been different if he was still around.
It seems that this model is not so common? At least the list on Wikipedia (https://en.wikipedia.org/wiki/Benevolent_dictator_for_life) only lists Django as an example with 2 BDFLs.
I was active in a smaller open source project (http://www.openlierox.net) where it mostly was like that, i.e. we were 3 main developers, and all major design questions were discussed among us. But this example is probably not so representative, as there rarely were further contributions by other people. But anyway, I was very happy with this model, that there is not a single leader, but 3 people.
And a project that starts with just 1 BDFL, and works fine & builds sth people "love", it'll continue working fine with that 1 BDFL.
If, however, replacing one of those 3, or adding another BDLF to a one person project — I'd think that by default, that won't work.
... Unless the new person has worked next to the earlier BDFLs for years, so the previous BDFLs can be ok certain s/he is "the right one"?
Whatever works fine, works fine — the risky thing, is introducing a change (a new person)?
Is forking no longer a solution to this problem? Or are you just wanting to wrestle control away from Linus?
Without the legal obligation to upstream, the majority wins what happens afterwards.
Why is that a problem?
Erhmmm, why does his age or colour matter in this context? Debate can happen about people and their leadership style, (I'm not a fan of Torvalds) but when the first criteria someone sees for criticizing another person is their age, colour, or race, that seems a bit shallow and biased.
> Racism: prejudice, discrimination, or antagonism directed against a person or people on the basis of their membership of a particular racial or ethnic group
I was then [regrettably] tempted to write something with a "mic drop" feel, just to express my disagreement in an acceptably pithy way.
And then it occurred to me that this whole sub-thread is Twitter-length comments. And I was about to add one more. And I realized that maybe the moderation of HN actually prevents us from going as deeply on this topic as we do on everything else, and so that moderator preference has essentially reproduced a pithy and unproductive Twitter dynamic for topics like this.
Pithy and shallow self-censored comments in reply to pithy self-censored comments.
EDIT: But also, moderation is really hard <3 Just wanted to try to put words to a small dynamic that's perhaps playing out here.
I'm not aware of any moderator preference against talking about this subject. Again: huge thread on it yesterday that persisted on the front page all day.
However, I don't think every single off-hand mention of race or gender needs to devolve into a subthread of 50 multi-paragraph comments. So I appreciate your decision to avoid it in this instance.
There's no such preference. What made you think there was?
The question of political topics on HN is a hard one because some political stories are on topic, some are off topic, and even the on-topic ones tend to turn into flamewars. I've written a lot about how we approach this: https://hn.algolia.com/?dateRange=all&page=0&prefix=false&qu.... Some good threads to start with might be https://news.ycombinator.com/item?id=21607844 and https://news.ycombinator.com/item?id=22902396. Also https://news.ycombinator.com/item?id=17014869, which shows how far back political discussion goes on HN, as well as the argument about politics on HN.
If anyone reads those explanations and still has a question that isn't answered there, I'd like to know what it is.
And redefining a word so is "bad for everyone but X" doesn't make X innocent of doing it, it makes X guilty and a bully.
Also, the world is much bigger than the US.