Apple Acquires FoundationDB
techcrunch.com
techcrunch.com
I don't understand why anyone would run a closed source database, especially with the open source options available.
What if startups with high switching costs (DBMS, OS, pretty much anything else infrastructural) added a guarantee to their sales contracts?
For example, it could be that, if they get acquired, they'll open-source the technology, sell it to another for-profit entity that will maintain it, or provide a migration tool.
Even for something like FoundationDB, it'd hardly be any skin off Apple's back to have a few employees spend a few months ensuring that previous customers have some sort of support.
This is what we are now doing for a project with Berkeley DB XML, which hadn't seen updates for five years. When there finally was an update, it was buggy and moved to the Affero GPL 3, which conflicts with other open source licenses used in the project. So, we continue to use the five year old iteration with a small set of patches.
(Lesson learned: once a product is owned by Oracle, prepare your evacuation plans.)
Fortunately it is stable (for my use cases), and it doesn't actually seem ton be that big a deal that it isn't being worked on. Django_REST_framework is a lot nicer though.
In the case of tastypie, I think all of the maintainers have switched jobs at least once in the last few years and at the same time the general Django community has been moving in the direction of simplicity rather than complex generic frameworks. Daniel's list of things he's not interested in implementing in restless is a good list of things which have been painful in tastypie: https://github.com/toastdriven/restless#anti-features
I've only been a minor contributor but I've increasingly found myself favoring really simple views – roughly https://docs.djangoproject.com/en/1.7/topics/class-based-vie... – since I work almost entirely on read-only public data.
If you or someone you know would like to work on tastypie, we're looking for new maintainers:
http://toastdriven.com/blog/2014/may/23/state-tastypie/
If you have an urgent patch, let me know & I'll see about merging it.
I have started using REST framework for other parts of the project, and it seems a lot more consistent with Django's other components (serializers are similar to forms, APIViews are similar to the generic views). In the end that just makes things easier. Its less context switching essentially, which is really useful when I don't touch that part of the project for a few months, then need to update something.
Thanks for the offer though (and thanks for the framework, it has been useful). And unfortunately I don't have the time to help with maintenance.
The typical large closed-source codebase is full of undocumented things and hidden dependencies on random chunks of proprietary environment. Making it usable as open source is a ton of work.
And you can't guarantee that work will happen without putting up some kind of bond or buying insurance, given that the vendor could simply go bankrupt and not honor the contract.
That would be like them putting a huge paper hat on their head, saying "I'm not a good target for acquisition".
Companies buy them so they (also) get their product/IP etc. If those startups have promised to give it away in such a case, then they are not that good of a buy.
That scenario won't hurt companies that are being acquired with the intent of keeping the product running, for obvious reasons. It also shouldn't hurt acqui-hire scenarios.
The only time it might hurt is when the acquirer wants to use the technology internally, but not offer it to anyone else. That's fairly rare, and the downside (potentially scaring away that tiny fraction of acquirers) is much smaller than the upside (making potential customers feel safe).
Often what is done is source escrow. The source is given to a third party, and if anything goes wrong the third party releases the source to the one who purchased the product.
Is it simply a case of the acquirer wanting the technology and maybe the personnel while the acquired company is losing money hand over fist so they just shut the company down?
The acquirer may love the technology and use it internally, but it's expensive to keep it in a public marketplace. It requires support staff, marketing, sales, etc.
There are also acquisitions that are purely about customers, so the startup's product is shut down or rolled into the acquirer's existing products.
So there are a lot of good reasons you'd want to company other than its product, but we can't always tell which reason it is right away.
"Technology" is rarely re-used per se.
I wonder how often this actually works. In most cases where I've seen this (or heard of it happening first hand from people I know) the vast majority of the people you'd want to keep were out the door of the new place pretty close to as soon as possible (meaning, as soon as the contracts allow, or the golden handcuffs are mostly off, or whatever is relevant to the specific situation).
On the inside there tends to be a pretty predictable path: New management tells everyone nothing will change materially, everything inevitably changes very quickly, people get disgruntled and take off for other opportunities (at a quickly accelerating pace as the old guard sees all their former colleagues from the old place leaving the new).
Like, don't build on top of freaking closed source platforms and after 17 years of OSI, 24 years of GPLv2 and myriads of open-source alternatives available, you'd think people would learn.
But yeah - Pretty much agree. The fact that most oss products are best-of-class makes the choice even easier.
a) look at building your own. What's inconceivable to one person may not be to another.
b) Build support for at least two closed-source alternatives, and make sure you always have at least two current alternatives.
Of course. Because Gimp does all the thing Photoshop does
Blender does all that Maya does I'm sure as well.
SCADA systems for industrial control? Just download the first one that shows up on Google.
EDA systems? Sure, there's a GPL alternative. People use Mentor (as an example) because they don't know better.
For most people, yes. Unfortunately Photoshop is pirated, so it isn't being used at face value.
> Blender does all that Maya does I'm sure as well
If you want to build your own stuff, which isn't uncommon at all and if you have a movie budget of 100 million dollars, you might be better off forking Blender. Also, Autodesk's Maya is not a startup and it isn't being used necessarily as a platform, so that's building a straw-man on your part ;-)
> SCADA systems for industrial control?
You mean Eclipse SCADA?
Funny thing is that I'm working on a software system using it right now. My mission is to steer and monitor power plants and nothing on the market was suitable for my needs. Building on top of an open source stack helped tremendously.
It actually does, for the majority of the population. Which doesn't matter. Any Photoshop alternative must be a clone of Photoshop.
Oracle and Teradata for example are proven databases with official support available worldwide and a talent pool you can draw from almost immediately. You don't get that with most open source databases (at least those that don't have a parent e.g. Datastax, Mongo, MySQL).
Enterprise companies will trade technology for stability and supportability. Most of us will flip the other way.
And I never said that people shouldn't use small startup technologies. Only that when you do you take the risk of the company not being around in a few years. And the people who will take that risk are really other startups or early adopters.
Correct. But you are betting the success of your nascent start up on another nascent start up. This is straight up wrong.
For a large company its different. They have all the resources to go into months long migration projects. As a start up, you can't afford time for migrations when you are busy doing the real work.
I strongly disagree your stance on Oracle in particular. I had two arrays at a medium size academic library 8 years ago. Anything I had to call about the data base meant a line item for business review due to cost if it wasn't covered by Oracle's service agreement.
PostgreSQL is amazing and I much rather work with that and hire whoever I want with what I want to do either per instance or annual contracts.
Which still sucks, since downvotes can lead to shadowbans.
PS. Just double checked the Guidelines, and this codification is no longer there. However, there's also no guidance to suggest an appropriate reason to downvote.
I believe I have a few invites.
I didn't have the time to delve through all the options out there for this purpose, and evaluate each of them, when there are out of the box solutions that were closer to my needs, though not strictly SQL based (Mongo, Rethink, ElasticSearch, Cassandra all come to mind). There is ~6k/month allocated to hosting costs, and ~$40k/month to the handful of people in the IT team... there isn't much wiggle room there for a small company, and everyone wears a couple of hats. The current application is using MS-SQL (hosted in Azure without redundancy) and MongoDB mirrored data for searching against... licensing to get a replicated MS-SQL setup for better availability would be more than our entire next generation hosting budget... If we could have actually talked to someone who wasn't a sales person at EnterpriseDB that could do more than send you a PDF sheet targetted at managers that might have swayed me.
Sorry, will end my mini rant.. in the end, what support I do have from MongoDB (using their backup service), and my experience with actually just using ElasticSearch and Cassandra has been far better for setting up for something resembling high availability/distributed configuration has been easier than even getting a proof of concept PostgreSQL setup working.
I really hope that PostgreSQL gets it together within the next year or so, it would have been my first choice had I been able to actually get some support within a reasonable budget for my needs, or if I actually had the equivalent of a DBAs salary or more to throw at the problem, which I didn't/don't.
I don't know the features of PostgreSQL that you want to use, but I'm totally willing to learn if somebody is going to cover my living costs. But I'm not even going to respond to your job ad if you put "PostgreSQL plv8 REQUIRED" in it.
For that matter, if you think it's simple enough for a part-time DBA, then why don't you just assign one of your existing IT people to learning and implementing the RDBMS that you need? Surely not all of them want to do the exact same job forever. PostgreSQL has excellent documentation.
If I was hiring a full time DBA, I would have put POSTGRESQL DBA as the job title, and made plv8 a feature requirement that I needed/wanted. As it is, there's no budget for that.
I've watched Oracle try to strangle more than one company once they were dependent on their database. One they succeeded, one managed to migrate to PostgreSQL just in time. If you build your business to be dependent on Oracle they have you by the balls; don't think they're not going to squeeze. And IME the worldwide talent pool is much more available and... well, talented, for PostgreSQL or MySQL or any of the major open-source options.
The cow was the target all along, not the ant.
I think it was Peeps.
http://www.ted.com/talks/dan_dennett_on_dangerous_memes?lang...
The parasites did their thing just fine for each individual organism. But that one random one that happened to cross wires in just the right way ended up doing orders of magnitude better. A rare occurrence, but very advantageous once it happens.
My theory is the parasite community for that species can survive other ways just fine, but has stumbled across a scenario that serendipidously gives it an extra boost at the change of reproducing.
I would imagine that most species that benefit from these complex interactions tend to have left-over backups for survival and reproduction. When species over-adapt to specific scenarios, it seems likely they'd become to susceptible extinction or their populations would be periodically thinned out.
That's one guess.
Berkeley's Understanding Evolution website has a nice overview of all the mechanisms of evolution. http://evolution.berkeley.edu/evolibrary/article/evo_14
It happens more often than you'd think. There are still many things in the database world that open source does relatively poorly compared to alternatives.
[0] http://stardog.com/ [1] They've started calling it a graph database, though I think triplestore is the most correct name
Presumably, because you have a business, which has a product, which has a nascent feature, which requires some particular set of time-and-space-and-distribution guarantees that no current database on the market makes. This is why, for example, Cassandra was developed.
Clark & Parsia had a history of open source (eg. Pellet) which was the best in-memory reasoner for a long time IMO... but not a lot of luck getting sustainable business subscription revenue. This led to the switch to dual-license AGPL in 2008 and now closed-source Stardog...
SPARQL should really be everyone first technology to investigate before heading off to anything else. i.e. when you are still pivoting every week you should have the most generic database tech possible. Only when you scale you should specialize.
How about FoundationDB? ;-)
-- sad production ex-user of Datomic
Would love to hear some honest feedback. Maybe your struggles were because of the tech, earlier versions, bad hardware config, or mis-applied use case?
Though it's not usually write throughput that most of these technologies are worried about. It's usually compression using dsp methods, aggregate stream folding computations, etc... that matters.
From an IoT or sensor network standpoint, 7000 writes/second is an idle server.
If you want a database for blazing fast data-storage and retrieval, there are many options available. You start seeing the real benefits of kdb+/q when you use q to simplify very complex operations that aren't easily done in SQL. Also, the high level operators that q makes your code extremely terse. I've written complex backtesting systems that perform data mining on massive datasets - all in one page of very tight q code!
I would love to find a fast, scalable open source db that implements Foundationdb's features.
EDIT: Also, I'm not sure there's production ready free software column-store DB.
Postgresql had it long before Oracle, but it was dropped as being too much of a hassle to maintain somewhere in the 7->8 transition IIRC.
As a thought experiment, if the company just went bankrupt and had no one working for it any more, you contract won't help right? Second part, what happens if someone just buys the assets (IP) and closes the company, then hires the old team? There are many possible variants here.
Good luck to all the former FDB'ers.
This is why everything you build your own product on and that can't be replaced in a matter of hours by a competing product should come with source. Worst-case scenario, you fix the issue in source yourself.
1. the new owners aren't interested because they want to product to cannibalise for something else they're doing. Original product will never see the light of day again, no new bug fixes and possibly no more security patches.
2. the support contract cost is hiked up from maybe a few hundred dollars a year to several thousands
or the possibility of:
3. one or other of above but with the product's support staff reduced, dispersed or fired.
#1+3 above happened to us after Oracle bought out a bunch of stuff we relied on heavily for our hosting platform. Was a very painful time.
I generally share this opinion. However, in this case no open source solution came close to the features offered by FoundationDB. There are a couple of attempts (like CockroachDB) which could achieve something similar in the future.
One more reason for me to hate Apple now.
if you think you've got the secret sauce, and you've actually had to put it into action and still hit five nines, awesome. But IME doing that with something like PostgreSQL is non-trivial (read: costly). That's why FoundationDB looked so appealing (to me).
It's been used in production by big ad houses like AppNexus as well as retailers like SnapDeal. Lots of miles on the code.
Was closed source for years, but went open source about 9 months ago.
VoltDB offers full transactions / open-source. Lots of differences between us and Aerospike / FDB.
You might be surprised how many apps deploy on VoltDB with cross-partition transactions making up a solid chunk of their workload. Yes, they're slower than partitioned operations, but they're still faster than MySQL much of the time.
Most of the apps we see partition very well, especially for writes. The fact that they can run 10k distributed aggregates a second to get a global view is something few other systems can touch.
You pay me and the licensing question goes away. You still get to poke around the source.
MongoDB, RethinkDB, ElasticSearch, Cassandra and I'm sure a number of others.. each of them have HA options (though RethinkDB is still a few months out for auto-failover iirc), and Cassandra has pretty close to linear scalability in production for some very large data loads. I really like each for different reasons, and would lean towards one or another depending on load.
Not to mention, my speculation is PostgreSQL will likely have an in-the-box replication with failover and/or multi-master solution in place within a few versions.
A natural fit is something like dropbox, evernote, feedly, etc. Every user has his own actor and it is a full SQL database. Which makes the kind of queries dropbox would need very easy to do.
It also has a KV store option. This is basically a large table across all your servers. This table can have foreign key sub tables. From my understanding of FoundationDB this part is closer to what they were offering.
This is not to rubbish it - I've not used it after all - but the claims being made for ActorDB are pretty far away from the claims made for Foundation.
ActorDB fits a specific data model extremely well. Some less so. But that is the case with all databases.
Agreed. And what I'm saying is that it appears that ActorDB's per-shard area concurrency is limited to one writer. And that means that SQLite's concurrency support is (contrary to your earlier post) extremely relevant: not just in terms of pure performance, but also ability to perform concurrent operations. If you need more concurrency, your only choice is to shard extremely heavily (which might mean you require more cross-shard operations, which are apparently slow).
As you say, some data models fit the actor model well, but this is still a far cry from the capabilities that were promised by FoundationDB.
The reason why I said sqlite concurrency support is irrelevant is because ActorDB serializes requests anyway. It must do so for safe replication.
Interesting - I didn't know that. Even so, it depends on what kind of writer concurrency we're talking about, I guess - I presume that ActorDB is limited not just by having to run requests one at a time per-process (which is a legitimate tactic to avoid latching overheads and so on), but by also not being able to run any new transactions against an actor that's received a write until that write commits?
> The reason why I said sqlite concurrency support is irrelevant is because ActorDB serializes requests anyway. It must do so for safe replication.
Do you mean by this that the entire cluster can only perform one request at a time? Or am I misreading you?
Read/write calls to an actor are sequential. I'm quite sure this is how other KV stores like Riak do as well. They have X units per server and those process requests sequentially. Their actual concurrency is basically how many units per server are running. They may interleave reads/writes per node or they may not.
ActorDB does not allow reads while a write is in progress. It is quite possible we will optimize this part in the future as it is quite doable.
But yeah, what made FoundationDB's SQL-Layer exciting (for me) was:
- For small clusters it was free. - Automatic HA - Operationally Inexpensive (talking admin time/effort/training; far cheaper than PostgreSQL) - Horizontal Scalability
The things that didn't matter for the 99%:
- It wasn't very fast.
I don't have "big data" problems (I could invent some). Most small shops (I suspect) don't.
The problem I do have is 3AM pagers, availability, wearing too many hats, putting dozens of hours into learning and experimentation to get PostgreSQL to use the hardware it's put on effectively, coming up with complex CARP+REPLICATION+FAILOVER plans, ZFS snapshotting because PostgreSQL still can't match the backup/restore process any commercial database had nailed down two decades ago, backing up the snapshots, figuring out how to partition clients into different table-spaces, blah blah blah.
You sacrifice some single-client performance with FoundationDB, but you solve almost every other problem you've got. And you now have the option of deploying a couple extra nodes to exceed your previously fairly intractable TPS milestones.
It's so easy in fact, you can now autoscale your database with your application servers.
And for your 80% of smaller clients it's absolutely free.
Such a bummer. :-(
This organization has no public repositories.
That is to say, people use Oracle, SQL server, Teradata, etc and they are not worried they will go out of business or be sold or otherwise that they will be left in the dark by a sudden shift in business practices and product availability.
The problem is almost entirely with startups, which are an easy target for bigger companies interested in their technology and team skills set. This is even more so an issue because the majority of startups are VC funded and are under pressure to sell or comply with VCs interests.
So in practice you have three ‘safe’ ways to build your infrastructure, which are not mutually exclusive. Choose OSS software, buy from companies such as Oracle, IBM, HANA, etc, and build it yourself. Depending on the expertise of your team and funds available for purchases, as well as qualifies of the available solutions, OSS is probably the safest way, followed by purchased from large corps. Rolling your own infrastructure is expensive, time consuming, requires committing your developers to building infrastructure instead of well, building a product, and may not work in the end at all.
Even the companies that can afford to do it everything in house (Facebook, Yahoo, Twitter, etc) choose OSS for the majority of their needs and build on top of that. Google is Google. However, if that infrastructure is your selling point and what differentiates you from the rest, and/or you have very specific needs and makes more sense to do it this way, doing it yourself can be a great alternative.
We rarely use OSS here, and we don’t use any proprietary infrastructure technology. We have built everything ourselves, and it has worked out great so far, but we have put a lot more effort and resources into that, whereas we could have instead invested on the Product. If we had to make a choice again, in retrospect, we ‘d most likely have gone the OSS way. It’s all about tradeoffs.
So, you don't use a libc, or a web framework, or any library that implements common logic that other programmers have spent ages refining?
If yes, erk, I don't want to be anywhere near that codebase. If not, I'm reminded of https://www.youtube.com/watch?v=YKjPI6no5ng
FOSS or proprietary to yourself are the only valid options.
Not that I agree with the premise that capitalism is what drives people to produce things to begin with.
Generally the people I've met who work hard to develop themselves and develop skills society needs end up doing quite well for themselves, although I admit that as an Australian my experience is probably different from the US. Here we have more socialised education and healthcare, so anyone with motivation can go to college.
http://en.wikipedia.org/wiki/Socio-economic_mobility_in_the_...
The good news is everyone is getting richer (rising tide), the bad news is inequality is increasing and low mobility is nothing to be proud of. Like logicchains I'm Aussie and we fare better at the moment, though we're also starting to head in the wrong direction by these measures.
Fortunately they are not very competent and do not have full control of the parliament, so most of their bills have been blocked by minor parties.
The not very competent people who wave snowballs around in congress, send bitchy letters to Iran and Israel, and squawk about #OBAMANET have full corporate sponsorship and the propaganda machine that is our media keeps getting these imbeciles elected.
The US is definitely treading water right now. I remain hopeful, but not optimistic.
* Destroying our earth with a pathetically broad plan for stopping (read: we won't).
* We're creating technological systems that pose the greatest threat against freedom ever.
* We're killing hundreds of thousands in the middle east because they pose threats to our Saudi oil fields.
Yay, productivity.
What I'm saying here, is that it doesn't have to be this way. Capitalism drives a population to productivity in the same way meth does: destructively.
I have no good answers for you. If the atrocities above bother you, you can do your part and opt out from the sides of society that requires you to be a part of it.
We just need a cultural shift to stop being such consumers. Stop buying a new phone every year, your current one can easily suit you for the next 10 years. Don't buy a new laptop. Start being cognizant of the influences brand names have on you and try to resist them where possible. Most importantly, we need to strengthen unions and support our local coops.
Start being aware of where the money you spend ultimately ends up.
</actual-advice>
<shill> I'm a libertarian, and I use that in the non-US definition of the word, which is to say I'm an anarcho-syndicalist. Unfortunately anarchism is widely regarded as unrealistic, but if it weren't for the Soviets mucking around in Spain in the 30's, it might be a very different story.
What is anarchism? Here's how Noam Chomsky, (the same Chomsky you know from your compilers / CS theory course) describes it:
""" Well, anarchism is, in my view, basically a kind of tendency in human thought which shows up in different forms in different circumstances, and has some leading characteristics.
Primarily it is a tendency that is suspicious and skeptical of domination, authority, and hierarchy. It seeks structures of hierarchy and domination in human life over the whole range, extending from, say, patriarchal families to, say, imperial systems, and it asks whether those systems are justified. It assumes that the burden of proof for anyone in a position of power and authority lies on them. Their authority is not self-justifying. They have to give a reason for it, a justification. And if they can’t justify that authority and power and control, which is the usual case, then the authority ought to be dismantled and replaced by something more free and just. And, as I understand it, anarchy is just that tendency. It takes different forms at different times.
Anarcho-syndicalism is a particular variety of anarchism which was concerned primarily, though not solely, but primarily with control over work, over the work place, over production. It took for granted that working people ought to control their own work, its conditions, [that] they ought to control the enterprises in which they work, along with communities, so they should be associated with one another in free associations, and … democracy of that kind should be the foundational elements of a more general free society. And then, you know, ideas are worked out about how exactly that should manifest itself, but I think that is the core of anarcho-syndicalist thinking. I mean it’s not at all the general image that you described — people running around the streets, you know, breaking store windows — but [anarcho-syndicalism] is a conception of a very organized society, but organized from below by direct participation at every level, with as little control and domination as is feasible, maybe none. """
One big misconception is that anarchism means that there should be no laws, and that murderers should be allowed to wander the streets. What a lot of people don't know is that anarchism, just like communism, was also a victim of the propaganda machine that we now call the red scare.
I think that a lot of people in tech, who can directly see how open source killed proprietary software, are the people who are most open to the idea this shift can happen.
Anyway, that's just my $0.02.
If you made it this far into my comment, give this a read: http://www.alternet.org/civil-liberties/noam-chomsky-kind-an...
And also read On Anarchism by Chomsky, it's fantastic. </shill>
That's a tacit acceptance that the solutions youre porposing have absolutely no chance of ever happening at a scale that will ever matter. Politics, and economics, are the art of the possible.
Capitalism coupled to represenatative democracy is the best shot w've got at developing a fair, balanced and sustainable economic system. What we need to do is correctly and rigorously price in environmental costs into the financial costs of our economic activities. Otherwise you get Soviet Russia laying waste to vast swathes of territory with misconceived development programs, or China polluting it's own country and population to death due to zero political accountability. The problem with Anarch-syndicalism is that at scale people will syndicalise back into special-interest blocs and you'll be back where you started.
Tend to agree, but that only addresses the economy. From what I've seen, capitalism cares very little for the advancement of society itself, and quite often works against it (eg, oil companies and global warming). You mentioned pricing in environmental costs, which obviously I would agree with, however I think the only way to achieve this is by banning companies from having any sort of political free speech and this is a tricky line to walk. In our current state, the environmental costs of our activities are a large point of contention because the people doing the damage are able to buy a large portion of "democratic" mindshare through propaganda. How do you regulate this? It's a hard problem.
Also, take something like consumerism (as in, buying a new phone every year). It doesn't make us happier, it doesn't make our lives better, nor does it do the planet we live on much good. However, consumerism and capitalism have grown into a feedback loop. It's an area where we spend enormous amounts of energy to derive very little benefit. The free market here does us no good, and in fact enables what I would consider a bad societal habit. Not that I think there's an easy fix or we should try to control people, but it's an example of capitalism working well but providing little value. It's self-referential existence.
> The problem with Anarch-syndicalism is that at scale people will syndicalise back into special-interest blocs
I am an anarchist, but I do believe this is true. Humans are not capable of this self-organizing yet.
I think the goal is unattainable at humanity's current level of spiritual development (which is, to be blunt, maybe a few millimeters further along than our ape cousins). That doesn't mean that the concepts can't be applied in every day life, however.
Living in the US, it's hard not to be completely disgusted by what passes as a "libertarian." I've grown to hate the word, and avoid most people who parrot it. The "less regulation!" "small government!" "free market!" drum gets beat all too often without addressing the elephant in the room: Corporate America is a wild beast running amok over the entire globe. The last thing it needs is less regulation! All the innovation and progress in the world won't be worth a damn if we're all breathing in toxic air and birthing flipper babies.
A lot of this is driven by the American culture's need for the new (as you pointed out). The sickness of our culture is the fuel of our economy, which is now built upon the backs of third world nations (which, by the way, are starting to equalize...soon there will be no more backs to climb on, what then?). We don't produce anything anymore, we just consume.
Even in SV, where people are constantly crowing about how innovative everything is, there's very little real, actual change happening. An app that deletes photos you send to someone after 15s. Amazingly innovative. Another chat app. Useful? Sure. Innovative? No. In fact, tying this back into the parent comments, I'd argue that almost all of the innovation I've seen comes from open source. 99% of private companies in the valley are doing some that has been done 1000x before, but just slightly better. It makes money, sure, but it doesn't help the world or advance society. Capitalism at its best.
I think there's a strong balance that needs to be struck between what actually advances society as a whole and what allows the individual to prosper. In the US, at least currently, the two seem pretty mutually exclusive.
/rant
The problem is breaking down the "rights" of corporations.
* eliminate corporate taxes * establish structures that allow corporations to hold on to underutilized/unutilized assets (follow through on this) * reduce intellectual property rights assigned to corporations * create a non-living entity legal classification with explicitely reduced rights * remove speach rights from corporate entities (employees, shareholders, etc still have those rights, companies don't) * restrict any propaganda spending by corporations
With those checks in place corporations can still exist, but would be geared towards growth (like Amazon) with continuous reinvestment, or towards paying dividends to those shareholders who are paying taxes.
With those checks in place, a basic/living wage and flat tax could be put in place, no loopholes, no tiered taxation.. everyone is taxed at 50%, everyone gets the same base wage check... the revenue is split between federal govts and state.. 25% to base wage, 25% to federal govt, 35% to states based on population, 15% to states based on land mass (perserve public lands).
Beyond any of this, there are way to utilize capitalism to serve the public interest.. just because this hasn't been done doesn't mean it can't be... and with appropriate checks in place (mainly in political finance which require the first steps outlined), stand a far better chance of succeeding than any alternative that has been tried.
The inherent problem in any case is concentration of power, whether it be in the state or in megacorps.
Economic freedom is a scale. On one side, there's 100% freedom of exchange/ownership, pure propertarian capitalism, which doesn't exist on Earth right now. On the other side, there's pure communal communism, which also doesn't exist on Earth right now. If neither extreme nor minimal economic freedom would address the issue, how could some intermediate level do any better?
Let alone the fact that we are only talking about economic freedom now, which is only one of many interacting facets of politics that can't really be isolated.
Allowing companies to buy any companies has issues. Not allowing companies to buy other companies also has issues. So how could any intermediate situation not have issues? If it forbids in some cases, it will have some of the problems associated with forbidding. If it allows in some cases, it will have some of the problems associated with lenience.
The argument I responded to was essentially asserting the existence of a non-capitalist solution that would have prevented Apple from buying FoundationDB without negative consequences. If I successfully argue that there are no perfect solutions, than that undermines the argument to which I was replying, and forces the poster to engage with the issue in more detail than just "capitalism is bad".
You seem to reason in really absolutist terms. You can have an open economy, where a government can still intervene if the current market situation has an extremely negative effect on society.
E.g. breaking cartels, monopolies or oligopolies where they seriously hurt a population does not throw away all the benefits of capitalism.
The disadvantage of one extreme is that you cannot have free enterprise, the other extreme is that you might end up with a few megacorps who control the market and ultimately society. In the middle you have a situation where there is free enterprise, but as a cooperation you also have to play by the rules that were set up to maintain fair competition and avoid centralisation of power.
The middle situation faces the potential of http://en.wikipedia.org/wiki/Regulatory_capture. This state-enforced monopoly is a small-scale manifestation of the complete state monopoly associated with 100% state ownership. The only way to completely avoid regulatory capture is to have no regulatory agencies, but this of course brings troubles of its own. There's no perfect middle.
Simple. Both extremes concentrate power eventually (in the state or in megacorps). Somewhere in the middle, you have an open and capitalist market, but the government keeps large cooperations in check, and the population keeps the government in check.
It can be argued that the EU strives to follow this model (although imperfectly). E.g. by enforcing net neutrality, being relatively active at breaking down cartels, regulating roaming costs (since the industry kept them artificially high), etc.
Of course, it's never perfect, because the circumstances are never perfect. So, you have to finetune and adapt.
What in this middle ground would prevent Apple from buying FoundationDB? I can't see this happening in any middle-ground countries like Europe.
What if there were no large corporations at all? What if IP and status/cash-flow were set up as the property of individuals and/or small teams who collaborated on a per-project basis?
You could have a system where IP was still shared in a completely open way, remaining free for non-commercial use, but commercial use would require a per-use payment, and commercial modification would attract a revenue share of its own if it was useful to a market.
This might not be ideal - it doesn't solve the problem of actually making stuff, for example. (There are possible answers to that, but they're even weirder.)
But it shows it's at least possible to begin to think about systems that don't have dinosaur corporations stomping around the ecosystem predating anyone and anything who's small and interesting.
And it specifically solves the problem of useful IP being removed or suppressed just because it can be.
The long term goals of that project seem to align.
It's the same with all software, really: to be able to do a large scale project you have to have funding from somewhere, as you need developers full time working on it to fix/write the stuff no-one wants to fix/write. Some OSS projects get this funding indirectly by sponsored developers who work at company X and write OSS code all day for the project (this is what Linux uses). Other projects are funded by VC money, licenses, support contracts or ads. It's not common a large scale OSS project is successful and stays successful without any funding from the outside.
Thus, if a piece of software gets its funding by selling licenses (like with Hyperdex Warp), it's the same as with a company getting funding by sponsoring: if the cashflow stops, the show ends. In the case of OSS you can grab the sourcecode at least, but in the end, to successfully maintain that it takes a lot of effort most of the time, as the projects are often large, complex and the internals unknown to the user.
Here's the architecture diagram for FDB, it's pretty fun to read: https://foundationdb.com/files/Architecture.pdf
See my other comment. https://news.ycombinator.com/item?id=9262673
Always something to be mindful of.
Also note that there seems to be a mirror for the PyPI packages.[3]
[1] https://pypi.python.org/pypi/foundationdb
Fortunately, they can't pull the artifacts from Maven Central.
Pulling an open-source project upon which people may depend is total jerk behavior.
It's out of date - only goes as far as 1.5, but at the very least you can import the 1.6.1 source as one big commit.
https://webcache.googleusercontent.com/search?q=cache:TllYak...
Of those four commits, one was a version bump, one a README tweak, and two were merge commits.
https://webcache.googleusercontent.com/search?q=cache:IzAL7H...
The source code is there on Maven Central, alongside the build artifacts, but the git history is gone.
Whenever Apple acquires anything that runs on a competing company's platform, that version is immediately killed (see any of their mobile app acquisitions).
Thanks for making things that much harder for every other database startup.
If you need evidence of Apple's ability to acquire companies while keeping product lines independent, look no further than Beats.
If you need evidence of Apple's ability to acquire industry leading technology to ensure exclusive advantage, look no further than AuthenTec (Touch ID).
If you need evidence of Apple's ability to sincerely invest in open source for the benefit of all, look no further than CUPS. Or LLVM. Or WebKit.
Apple has no pattern.
Knowing their past behavior would be an interesting indicator of what is most likely to follow here.
According to the TC article (and their website), FDB is no longer available for download and there is a "goodbye"-esque type of message on their community site [0]. Their github [1] repos have all now been made private. This seems most unfortunate for anyone who's included FDB in their tech stack.
For reference, FoundationDB has been compared to Google's F1 database [2] [3], so this is Apple purchasing a pretty shiny piece of technology.
[0] http://community.foundationdb.com/index.html
[1] https://github.com/FoundationDB/
[2] http://blog.foundationdb.com/7-things-that-make-google-f1-an...
[3] F1: A Distributed SQL Database That Scales - http://research.google.com/pubs/pub41344.html
Edit: More recently, they killed off OpenNI about a year after buying PrimeSense
CUPS was bought by Apple in 2007 and continues to have open source releases and continues to be used by major Linux distros.
For FoundationDB, there isn't really the same market. Most sites don't need a distributed database. People that run sites is already niche and people that run sites trafficked enough to require a distributed database is even more niche. I use HBase and Kafka a lot and there are definitely high-profile users of the software, but it's nowhere near the number of users of MySQL or PostgreSQL.
Maybe Apple will look at FoundationDB as something where open-source in no way hurts them and gets them free development. Google won't be replacing F1/Spanner with it so it's not much help to what a lot of people would consider Apple's largest competition. Plus, open-source could help make it better.
I wonder if a similar architecture could be used for building a distributed database that could rival what was lost with foundation.
The key with transactions is that they make life easy for the client. AP systems are easy for database engineers to write (and yet they still manage to screw them up :P), but systems that support ACID constraints are easy for application developers to use. That's why FoundationDB was so special; they promised the best of both worlds; the horizontal scaling of traditional NoSQL systems, with the ease of use (w.r.t. reasoning about concurrency) of SQL/ACID systems.
An oracle replacement won't be built for financial services companies.
If you want those same features, consider VoltDB. It makes different tradeoffs on what kinds of ops are fast, but has all of those features.
Essentially arbitrary two-key transactions are slower. Partition-friendly writes and global reads are faster. Global reads with real SQL can be orders of magnitude faster.
You probably meant [a-zA-Z]+DB!
I also can't help but wonder how much my options would have been worth.
edit: scratch that, $23 million is the amount they raised, not the amount of their sale. Regardless, unless they sold for 10 figures, I don't have any regrets.
Quoting the Google F1 Paper: "Features like indexes and ad hoc query are not just nice to have, but absolute requirements for our business."
Cassandra got ahead of this with CQL. FoundationDB saw this coming and bought Akiban to add a SQL layer. But bolting SQL onto a KV store, even a really good one, isn't trivial to do. I'm not sure it ever delivered on the promise of a real query layer.
Still, I hope this is a good exit for the FDB team. The KV layer is pretty cool stuff.
Full disclosure: VoltDB engineer here.
Sure, but isn't that basically what everyone does? All of the relational databases I know use ordered key-value storage engines. FoundationDB is the same, except it's distributed.
The point is to use FDB as a foundation. I heard how one of the FDB founders replaced SQLite's B-tree storage engine with FoundationDB, which turned SQLite into a distributed SQL database. It's much more difficult to make something like that if you had to tackle the hard distributed systems problems.
(I interned for FoundationDB a couple of years ago.)
> The point is to use FDB as a foundation. I heard how one of the FDB founders replaced SQLite's B-tree storage engine with FoundationDB, which turned SQLite into a distributed SQL database. It's much more difficult to make something like that if you had to tackle the hard distributed systems problems.
So this is a really interesting point and I can see why it makes sense if you’ve not built a SQL engine. So why is building SQL on top of a KV store the wrong call? What’s the difference between MySQL’s or Postgres’s or VoltDB’s “storage engine” and what FDB had built?
First, I’m not claiming that putting SQL on top of a KV store like FDB is impossible, just that you’re going to have to either compromise the purity of the KV engine substantially, or you’re going to get slow SQL for anything other than single row CRUD.
It starts with metadata. FDB-SQL stores metadata in the KV store itself. This is great from a distributed correctness point-of-view. It’s also great from a simplicity point-of-view. If I trust the underlying system to be safe and consistent, then my metadata is also safe and consistent. But now I need to do a ton of reads before I run my SQL to know how to run my SQL. Where’s my data, for example? Compare this to VoltDB which replicates metadata to all processing sites, and basically has a second state-machine to ensure that each processing site has the right metadata at the right time. Updating metadata is more expensive, but the fast-path of using the metadata is several orders of magnitude faster.
Then you get to locking. I believe FDB-KV uses per-key read-write-sets to manage concurrency. This choice makes their two-key transaction benchmarks look great, but it scales poorly as the number of keys per transaction grows. SQL basically begs users to write operations that read and write to lots of keys. To make scans fast (even common partial-index-scans), you’re going to need more granular locks, or you’re going to need to relax consistency (ACID, not CAP).
Now we get to storage efficiency. If I have a SQL relation with a primary key and other columns, do I split it so the primary key is in the “key” and the other columns are in the “value” in the KV store? Do I store the whole record in the “value” and loose some efficiency? What about relations with no primary key? Say I’ve got a table that facilitates a many-to-many relationship and it’s just a set of integer pairs. How do I store that efficiently in a KV store? And what about that pesky metadata? Does the key need to include the identifier of the SQL relation (table)? Of course it does.
Secondary indexes. Oh man. To do them well on top of a KV store, you’re going to need pretty strong consistency to ensure that index records point to the right base record, and that there are no base records without an index record. This rules out the aforementioned trick of relaxing consistency to get faster many-key transactions. It’s also a metadata problem; I’ve got metadata I need to read to understand how to use the secondary index. More than that, when I update a record in the base relation, I’m going to need to find all affected secondary indexes and make sure they reflect the new information. That might mean lots of reads to the system to query metadata to update the secondary indexes. And again, we have the efficiency problem where I have to put a bit of extra stuff in each “key” to identify the secondary index the KV-pair belongs to. So secondary indexes have locking/consistency, storage efficiency and metadata problems in this model.
Finally you get data transfer. This is probably not a broadly applicable problem, but I think in the FDB implementation, there is a lot of needless data transfer between KV cluster nodes and SQL processing nodes. Too much data needs to be collected and processed, rather than pushing down that processing to the data’s resting location. In FDB-SQL, if I just want one value from a large record, do I have to move the whole record over the network? Most SQL systems build processing DAGs for each SQL statement and can in-process stream between nodes, or have efficient temp tables to buffer intermediate results.
This falls out of making the SQL layer so separate. In fact, I think the team working on SQL was almost a completely separate team in Massachusetts than the KV team in Virginia. If the SQL layer worked, then I can see how awesome this would be. The “Layers” model just sounds so appealing. It’s just technically quite hard.
So yes, I agree that under most SQL-relational systems, there is a storage engine that smells much like a KV store. Still, that system is much less “pure” than what they were going for at FDB. Locking, metadata, secondary indexes, native understanding of relations, moving the processing to the data — all critical to do right to get reasonable performance.
If FDB had more time to continue on the SQL path, I imagine it would dictate much deeper hooks into the KV side of things. I’m sure with time it could get a lot better. There’s a lot of research addressing some of the tradeoffs I’ve mentioned above, especially in the distributed transaction coordination space. Still, it’s always going to be easier to build the storage engine for the kinds of operations you want from day one.
From what I know Apple is a big user of Cassandra and Teradata for iCloud, iTunes etc. Both of which are very solid databases that have been proven to scale.
I am not doubting FoundationDB's credentials but it's pretty extraordinary if they are having scalability issues with either.
My guess is that Apple plans to create an equivalent to Facebook's Parse. Either that or this is an acquihire.
Still a success for the team and should be celebrated appropriately. :) Startups are hard.
Apple used to claim to fight against Big Brother, today they are largely seen as Big Brother.
Thank god for Android for keeping them honest.
1. Why waste resources in maintaining something you don't intend to continue to support.
2. Given the litigious nature of the US, Apple probably doesn't want to deal with a codebase that could include all kinds of potential legal issues that are now suddenly Apples problem.
http://blog.foundationdb.com/databases-at-14.4mhz
It's a pretty impressive number, though as always with benchmarks, should be taken with care.
Nothing but good things to say, other than their multi-node configuration being a little strange.
In any case, given that narrow feature set, I can't think of many production quality NoSQL databases with that feature set, and I can't think of any that replicated the millions of writes/sec on GCE like Foundation did, so Foundation seems to be quite special.
Interestingly enough Apple is a heavy user of Cassandra, so if this related to iCloud I wonder if they decided to replace it and why. Or maybe they had enough money where the decided they could just own the database, and DataStax's valuation was too high.
EDITED: seancribbs corrected my acronym spelling in the early morning; *CA - DOES NOT EXIST, what I wrote initially was completely wrong
CA doesn't exist.
I'm not sure that's a good way of putting it. You can have databases that offer both consistency and availability, but they need to fit on a single machine (no partition tolerance).
Garbage collection requires (a delay that is indistinguishable from) partition tolerance. It's pretty illusory on single machines as well. Especially if they're multi-core.
And now they have been subsumed by Apple. "and they were never heard from again"
EDIT: a while back I commented on a comparison of FoundationDB to Cassandra, and Dave Rosenthal, the FoundationDB CEO, took the time to respond. I completely disagreed with him at the time, but think I understand his points now. Link: http://news.ycombinator.com/item?id=8745462
Full ACID and looks to have support for FK/PK JOINs/ and multi-table queries. Their benchmarks page looks super awesome. How well does this work in practice?
One major difference from Cassandra (I think) seems to be that coordinator nodes are set statically (but can be changed on the fly). Cassandra is not this way. While there are coordinator nodes during client connections, they are chosen dynamically by the client during the session and not some fixed config point.
Another difference appears to be transaction limits (https://foundationdb.com/layers/sql/documentation/Concepts/k...). This is fundamentally different from Cassandra, but to be expected without tunable consistency.
Different tools for different problems. I think Cassandra fits Apple's content distribution model better (e.g. streaming music/movie blobs out of C* for iTunes all over the world), but for a traditional RDBMS that is distributed, this looks like a great escape.
Scaling seems pretty easy if it's as easy as copy-pasting the config file from node to node and bouncing the service. Anyone know about this in practice?
“The most important innovation with MongoDB is its API,” Rosenthal said. “We sell an amazing storage technology that could be compatible with NoSQL technologies like MongoDB.”
While you could point to the pulling of downloads, and pulling of GitHub accounts as a transitional 'rebranding', I just don't see what incentive Apple has to open source a system that is useful to bigger players and generates a lot of revenue.
I'm sure a company that's richer than many countries is really, really concerned about the cost of development.
The benefit of making sure no-one else can benefit from the technology, on the other hand...
And you're spot on about the cost of development. I'd imagine 12 - 48 hours of profits from Apple would be enough to fund its database R&D for the next 10 years.
Clearly doing this with ascii characters on a terminal screen is much easier than trying to render 3D avatars in the real world but the number of database transactions per second becomes massive. If we posit that there are 1,000 players visible to the current player, and just the player / world select is a single transaction, then at 30 fps, that is 60,000 or so queries per second, you throw actions between players, and between multiple players (area of effect attacks) and add another 2,000 monsters and you're easily over 250,000 transactions per second.
Since FoundationDB was talking millions of transactions per second with full ACID it seemed like it would be a useful back end for this. Or put another way, you could contemplate building this and see if their product was actually able to keep up at this rate. And they asserted this was on modest hardware (like 32 nodes).
So yes, I was looking forward to the interesting places that investigation would lead and the things I could learn.
http://en.wikipedia.org/wiki/Shake_%28software%29
"Existing maintenance program subscribers had the option to license the Shake source code for $50,000 USD."
This also helps solve the larger problem of concentration of wealth/power within large corporations as the cooperatives would be distributing the wealth to all participants and not just to founders.
It also provides a better path for startups so they don't have to borrow VC money and can get the help of the community to move them toward profitability in the long term rather than taking the first available bag of money from disinterested investors who only want a return on "their" money.
Cooperatives are among the most stable forms of enterprise and are run via direct democracy (at least in our view.)
To discuss this further, join our mailing list here:
groups.google.com/d/forum/coopsource
I also wonder if it is lawful to remove an opensource project overnight from public access, to which people outside the company might have contributed. I have not used this product before and I know neither its licence nor the amount or existence of its contributors, but if I have had made a significant contribution, I would be very annoyed in this case and have sued--if possible--Apple or whoever responsible for publication of the sources until the acquisition.
I'd love to hear from anyone who was actually using FoundationDB in development or production, and what other NoSQL open source alternatives you are considering. pcorless@aerospike.com ; http://goo.gl/KVQxyq
Does anyone know of any good (and proven) alternatives to FoundationDB?
Did any of you get bitten?
"Thank you for your support of FoundationDB over the last five years. We’re grateful to have shared our vision of building the best database software and we strongly value your participation in this community. We have made the decision to evolve our company mission and, as of today, we will no longer offer downloads. If you have any technical questions, please email info@foundationdb.com."
it affects sorting IIRC.
And in the case of your comment it makes it stand out as less readable as a sign for anyone who hasn't picked up the house rules yet. : )
I haven't really thought about this, but here are some changes that I would borrow from here and there with some that I haven't seen elsewhere:
- a downvote costs you something (a karma point or half a point) ala StackExchange, which makes you think twice before you're urged to punish;
- categorize votes ala BuzzFeed reactions insteading mindlessly up-/downvoting;
- your karma gets a share of the collective karma of your reply tree - those who start vivid discussions should be rewarded;
- reward with own karma - if you really like somebody's comment, why not donate some of your own karma and reward quality comments with more than 1 point;
- there's no point to keep punishing somebody for their multiple replies - this silences voices; you should be able to pick and rate the overall participation of somebody in the thread, not punish them multiple times for each attempt for them to convey the same thing.
Hey, here's a question, why does the comment reply link sometimes not render until I refresh the page? Are they on a timer or something?
How can a -4 comment be "noise" if there are tons of replies? I often have a root comment with -4 and then tens and at times hundreds of comments underneath?
And I agree with you regarding the real world karma consequences of this for HN :)
[0] https://lobste.rs/ [1] http://monocle.io/
(I've detached this subthread and flagged it as off-topic.)
FoundationDB was closed source, they just allowed to use their product for free in small deployments.
By the way, if you did not see it there was an update posted that states that FDB was closed source. It also doesn't matter that some components were open source or not, because without FDB they are useless either way.