OpenSolaris is dead
blogs.everycity.co.uk
blogs.everycity.co.uk
They aren't hackers. They aren't designers. They aren't entrepreneurs. They're business men.
Similarly if you dare compete with Oracle, you're going up against a company similar to Microsoft 10 years ago: rich and mean. Look at someone like Google, you think they're evil? Oracle could literally wipe them off the planet if they were inclined. The fact people hate upon Apple or Google, or even Microsoft when Oracle is the towering nemesis is hilarious. At least Oracle doesn't bullshit you. Google? Do no evil? You're a public company.
Vision, for the greater good, wanting to change the world are all well and great, but when the chips are down, you know what counts? Revenues. As Larry Ellison once said about Sun: "Lots and lots of blogs does not replace lots and lots of sales"
and you know what? I admire them for that. I respect that they walk that line. They're a public company, and god dammit, they're going to win. I would rather spend a year working with Larry than Steve Jobs, Bill Gates or the Google guys.
If Oracle continues with this kind of ruthlessness, they may squeeze several million out of Google et al, but Java is done for. Nobody will want to use that platform anymore because it will be too risky legally. It'd be a huge opening for .NET and other competitors. Java would go the way of COBOL, only running in the darkest abysses of megacorporations that never modernize.
You can be happy that Ellison doesn't play this "good for developers" thing, but what's good for the goose is good for the gander. By promoting software development and Java development specifically, Oracle would be ensuring generations of Java programmers, and lots of Java-related services and licenses from enterprises. If someone without any programming experience learns Java to write code for Dalvik/Android, as they may because Android is a "cool" platform that would attract young programmers, they can and likely will go on to big corporate environments and continue to work in Java.
They're going to be doing so for decades to come, either from Oracle itself, or someone who's paying Oracle truckloads of licensing fees to be allowed to publish their own Java implementation.
The only thing that could upset their subscription to money would be some inconvenient little company that decides to make a clean-room Java implementation that could target the same customers. This is what Google has done.
If they can win this in court or force a settlement that pretty much guarantees that anyone thinking of building that business model won't even bother to get off the ground, and Oracle can keep printing money.
It's funny that you should mention COBOL. Java turning into the next COBOL would be awesome for Oracle. IBM still makes billions selling systems to run COBOL[1], that's more than Sun ever made off Java.
Mainframes promise turnkey-reliability, where the IBM support guy comes in before you notice anything going wrong. The point is that "mainframe" is a strong brand and IBM has the monopoly on it. There are solutions magnitudes cheaper and just as reliable as a mainframe.
The message Oracle wants to send is pretty clear: if you want to use Java in the mobile and enterprise space, you're going to have to buy one of our very nice support contracts. Considering that this is par for the course for most enterprise platforms and mobile (until Google wrote Dalvik), why would previous licensees make the costly decision to switch?
That's the result of their bonus policies. If your bonuses reflect short-term earnings, that's the kind of earning your company will generate.
OTOH, this and the Google lawsuit are so completely boneheaded, I would not be surprised the people responsible for it end up like Rick Belluzzo or Sandeep Gupta - in nice positions at Microsoft properties. Even the lawyers are the straight out of the SCO lawsuit...
just trying to state that (IMHO) Oracle is a very different beast to Google, MSFT, et all.
Google: we're nice guys, don't censor china!
Oracle: we've been around since 1977. we've been to war before, and we'll win again.
Also the fallout could hurt Google far more than Oracle regarding market cap from public perception. I think people entrench enterprise stuff like Oracle as brutal, and Google as fluffy. IMHO.
Oracle has a $113B market cap. google has a 155b market cap. granted, both those numbers are large, but Google's is significantly larger.
how many google Engineers do you think it would take to build a rdbms that is significantly better than Oracle's product? It's well within Google's capabilities. And yeah, it'd be a patent shitstorm, but it's not like google lacks patents or lawyers.
I mean, granted, oracle is primarily a sales company, so simply coming out with something better, for free, won't kill them overnight. But it would significantly hurt them.
Could Oracle write a search engine? No. the idea is just silly. so Oracle would be limited to legal maneuvering in their attempt to overcome google. I just don't see it happening.
The leadership. And who they are.
For example, Eric Schmidt is apparently a super nice guy [1] Larry Ellison on the other hand, has cited (I believe, Ghengis Khan?) "It is not enough that I succeed, everyone else must fail"
So the leadership is drastically different in terms of how they work. Ellison is marine corp, Schmidt is peace corp.
Could Google make a rdbms that is significantly better than Oracle? On paper, sure. But Google isn't doing a great job (yet) at obliterating markets that are entrenched (Sharepoint, Outlook, Exchange, Facebook) so it's unlikely Google has the willing (due to leadership) or management (due to internal reasons, I guess) to crush Oracle.
I think if Oracle wanted to squeeze Google, they probably could actually. Google has made more and more enemies over the past few years, and I think if someone such as Ellison stepped in and decided to lead the blood hounds, it could be interesting.
Obviously this is all hilarious speculation, but it's pretty interesting.
See, while I think that would make it /much/ harder for a little guy to go up against ellison than agains schmidt, being as oracle is /smaller/ than google, I don't think aggression is enough to tip the scales.
the thing is, aggression only helps you if you have the ability to take down your enemy. I don't think Oracle has the ability to threaten google's core revenue generating stuff, search, adwords and adsense.
Oracle is an enterprise software company, and google is a consumer software company. Completely different markets. I think either one will do poorly breaking in to the other's market.
my point with the rdbms is that there's something relatively simple (well, relatively simple if you have Google's Engineering resources) google could do to hurt oracle. Now, obviously, for it to really hurt Oracle, it'd need to be marketed by companies that can market to the sort of people who spens money on Oracle, e.g., not google. but, if you release something awesome under a loose enough license, that's what happens. Other people will take the tech, relabel it, and start selling it. Look at all the companies that took ZFS and started selling network-attached storage devices. (granted NetApp is now using it's patents in an attempt to stop that sort of thing... like I said, there would be a legal shitstorm... but it is cutting in to NetApps core market.)
The interesting thing about just releasing a RDBMS is that it's a not-very aggressive, very nerdy way to get revenge. it doesn't require Schmidt to talk about blood or anything, just releasing some really cool new software.
Now, it would be really hard to actually kill oracle, but hurt it a little, in it's core business? yeah, I think Google releasing an open-source, loosely-licensed Oracle-quality and scale RDBMS under a permissive license would slowly do just that. Hell, I've personally watched managers debate Oracle vs MySQL... and many of those debates were close. many more would have ended up on the MySQL side if Oracle didn't have some significant technical and reliability advantages.
My understanding is that Google's core business and nearly all their revinue comes from adwords, adsense, and search. My understanding is that everything else is just fucking around- e.g. gmail doesn't make them money (this may be wrong these days, but gmail is certainly small potatoes compared to adwords and adsense.) There's no way Oracle could put a dent in google's dominance in that area. Just like google wouldn't be able to market to enterprises like Oracle does, oracle doesn't know how to sell to consumers, technical barriers aside... and for Oracle, the technical barriers would be enormous. Oracle is used to customers willing to go spend an order of magnitude or two over the cost of commodity disk on storage. In the search market, you have to very carefully write your software so that it can work okay even though it is running on the cheapest and least reliable hardware you can find. (One of my past jobs was dealing with the flaky hardware at one of the larger search clusters.)
I guess my main point is that I can see how google could contribute to others chipping away at Oracle's core business... I don't see how Oracle could contribute to anyone chipping away at Google's core business.
if that's true, then yeah, google is in trouble. I'm certainly not qualified to speculate about legal issues.
Still, the question in my mind would be: "how many of those patents would stand up to determined scrutiny from a well-funded and very motivated team of competent Engineers and Lawyers?
Market cap doesn't translate into assets that can be used in a fight. Some more relevant comparisons:
Oracle: 115000 employees, Google: 22000 employees.
Oracle: $27 billion revenue, Google $24 billion revenue.
Oracle: $9 billion operating income, Google $8 billion operating income.
Oracle: $60 billion total assets, Google $40 billion total assets.
> Could Oracle write a search engine? No. the idea is just silly.
Considering that a single determined programmer can make a great search engine (duckduckgo.com), I find it hard to believe that among all of those 115000 employees, Oracle would not be able to find a few who could make a good search engine.
yes, but Oracle is /very/ sales and corporate support (sales) heavy. those people will not help you build a search engine, at least not a search engine with Google's target customers and business model.
Now, 'enterprise search' for internal stuff? Oracle could do that. Oracle might even be able to get corps to pay for a subscription-based 'search for corporations' that cleaned up most of the porn and more shady spam. but they couldn't do an ad-based search targeted at everyone, like google does.
If nothing else, most people are aware that google knows everything about you, and uses that knowledge to serve you ads. this only works because people, by and large, trust google to not go too far overboard.
>Considering that a single determined programmer can make a great search engine (duckduckgo.com), I find it hard to believe that among all of those 115000 employees, Oracle would not be able to find a few who could make a good search engine.
do you think duck duck go will last after it becomes worth spammer time to game them? I've said before that the search engine market has something of a negative barrier to entry; your job is /much/ easier while you are small enough that most spammers ignore you.
Another problem oracle would have is that Oracle isn't cheap. the are used to environments where you can shell out one or two orders of magnitude more than you would for commodity disk to get a really nice storage unit. Running a search engine is all about making the cheapest hardware that could possibly work run without going down when a bit of hardware crashes in a weird way.
Overcoming this cultural bias towards "buy good stuff" is pretty difficult. Then, they will have to face a set of technical challenges they are not used to facing. There are a whole lot of problems that go away when you decide to pay for a really high quality storage unit. I mean it's so bad that I pay 2x as much per gigabyte so I can have "enterprise' sata and not worry about most of those problems. Judging from, say, s3 prices (at scale, its $0.037 per gigabyte for four copies of the data, if my information is correct, which it might not be) it's quite possible that the big players use hard drives that are even more prone to failure than the consumer-grade drives that I have access to.
"We will determine a simple, cost-effective means of getting enterprise users of prior OpenSolaris binary releases to migrate to S11 Express." -- Says enough.
"We will continue to grow a vibrant developer and system administrator community for Solaris." -- I really wonder how.
They really killed the wrong Solaris. If they embraced the GNU userland (as OpenSolaris did) they would have had a much better shot at getting more raised-on-Linux admins and developers to embrace their OS. As it is, the current BSD flavored Solaris tools are so crufty and incompatible it leaves many Linux users with a distinct loathing for the OS.
Companies are much less likely to embrace GNU anything, GNU just doesn't fly that way. BSD license is much more permissive, and therefore much more attractive to companies.
Oracle took a very clear stand about how they intend to generate value, one I cannot agree with and that I will not enable. I am an enthusiast of open-source and free-software, of value created by sharing and openness instead of ruthless competition and secrecy.
I was considering the use of OpenSolaris in a production environment, mostly due to ZFS. Now I won't. Maybe I'll think again about ZFS if and when FreeBSD takes whatever they can and fork it into a filesystem and toolset that's useful for me. These developments go completely against what I want and what I believe.
Enjoy your legacy products and your short-term profits, Larry. It won't last long.
We will continue to use the CDDL license statement in nearly all Solaris source code files. We will not remove the CDDL from any files in Solaris to which it already applies, and new source code files that are created will follow the current policy regarding applying the CDDL (simply, that usr/src files will have the CDDL, and the very small minority of files in usr/closed might not have it). Use of other open licenses in non-ON consolidations (e.g. GPL in the Desktop area) will also continue. As before, requests to change the license associated with source code are case-by-case decisions.
We will distribute updates to approved CDDL or other open source- licensed code following full releases of our enterprise Solaris operating system. In this manner, new technology innovations will show up in our releases before anywhere else. We will no longer distribute source code for the entirety of the Solaris operating system in real-time while it is developed, on a nightly basis.
In other words, the Solaris development process is becoming more closed, but you'll still be able to see the source code for any given release. Or at least, that's how I interpret the above.
I'm not in touch enough with Solaris development to know how much of a practical impact this will have--how many non-Sun/Oracle people work on OpenSolaris?
Oracle is a for-profit entity, they will take any legal or tactical advantage they can get (e.g., lawsuit over Java in Android), and they make no bones about it.
I don't like everything they do, but you know what you get with them.
Google says "don't be evil", yet they back off that principle when it's inconvenient (e.g., caving in to Chinese censorship demands, deal with Verizon over net neutrality, etc.).
Because that's what it sounds like to me. Personally I admire companies that are giving me value.
> yet they back off that principle when it's inconvenient
Regarding both net neutrality and Chinese censorship, Google did what nobody else did.
So I really don't get this line of reasoning.
Personally I take away from this that my company shouldn't give a crap about anything other than money, because all you're achievements don't mean squat in the end apparently.
Neither has given me anything that I didn't pay for, either directly or indirectly.
Regarding both net neutrality and Chinese censorship, Google did what nobody else did.
Which is what, exactly?
When push came to shove, they acted no differently than any other large corporation.
Personally I take away from this that my company shouldn't give a crap about anything other than money, because all you're [sic] achievements don't mean squat in the end apparently.
I don't understand your emotional attachment to Google.
mechanical_fish said it better than I could, so I'll just quote part of his reply here:
It would be refreshing if people stopped pretending that these inter-corporate battles had anything to do with either morality or the public good.
That's not strictly true. I would argue that Google provides a greater amount of consumer surplus than Oracle. Consumer surplus pretty much amounts to value you obtain from a product or service that the producer can't capture in the price.
http://techcrunch.com/2010/08/13/google-net-neutrality/
I am not exactly a starry-eyed admirer of the Gordon Gekkos of the world, but I tend to agree. It would be refreshing if people stopped pretending that these inter-corporate battles had anything to do with either morality or the public good.
Also, Oracle has never created anything really significant (in my opinion) other than their overpriced "enterprise" databases. Now it seems that they're hell bent on taking everything good that came out of Sun and destroying it.
I don't see either censorship in China or the net neutrality proposal (which I don't know much about) as bonafide "evil". You may disagree, but it's hard to say that they are evil or done with evil intent.
Oracle, on the other hand, is pretty easy to call evil because they act out of cruel self-interest practically exclusively. Oracle is always taking cheap shots.
I am highly skeptical of this. I believe this was the stated strategy of many of Sun's decisions, and it did not work out well for them.
In my opinion Sun's downfall was a serious overestimation of SPARC's performance. They needed to move and get something else out there.
- What did they get for MySQL? Nobody was buying Sun kit to run MySQL. There was no move into the LAMP space by Sun.
- VirtualBox? A free desktop-class VM which still hasn't been integrated with any other product or seemingly used to build anything.
Who pays for IDEs any more?
RMS would be a clear counterexample. (See Linux, GNU utils, the recent GPLv3 release of Castle Wolfenstein, and many many other victories.) It's a curious sort of logic that says a hardline stance is ineffective because 100% of his goals aren't meant.
Not that moderates don't also serve a purpose, but the radicals are an absolute necessity.
But nobody takes RMS's advice as practical. It's just too extreme. Most people don't believe as RMS stated in his recent Reddit interview that the only way to ethically use a pacemaker or other life-saving device that relies on proprietary software would be to start developing an open-source replacement.
RMS has been somewhat successful, but I think much less than he would have been if he hadn't poured so many resources into the silly points that should have easily been let go, like "GNU/Linux".
I'll just make a quote here :
The reasonable man adapts himself to the world; the unreasonable one persists in trying to adapt the world to himself. Therefore, all progress depends on the unreasonable man. George Bernard Shaw
If I were you, I think I would die from cynicism toxicity shock.
Well, either that or you think that all industry should be destroyed because all corporations should invariably end up being perfectly evil.
My comment was less an expression of admiration for Oracle than disdain for Google.
House is far and away the most popular character on the show, because, while he is an unpleasant, malicious, manipulative sod, he wants to figure out the cure to the disease, and make the patient feel good by being well.
Well, he's also popular because he's Hugh FREAKING Laurie, but that's another story...
This has become a bit of a meme lately, and it's pretty tenuous.
Google makes lots of money by not being evil. People have given Google tremendous data that no other could get because they have a general trust that it will be used responsibly. In all of the fawning over the just released voice recognition in Android, almost no one noted that it means that much more of your interactions with your device will be sent to Google, which is something that few companies could get away with.
So seriously, people need to shitcan the "they're a public company" or "they're out to make money therefore they are evil" noise. It's dumb. Oracle is essentially pissing on all of the Sun assets that they acquired for a short term game because there is no long game in it -- Solaris is on the way outs, and even Java is almost dead in the mobile space (not counting Android...), and in the Enterprise space despite all of the rhetoric on here, it's dying and is largely the domain of legacy projects. Oracle is cashing in while they can like a standard troll.
Google, however, is in for the long haul, and their entire strategy depends upon a lot of trust by consumers. In other words they MAKE MONEY, fulfilling their public mandate, by not being evil. Strange, isn't that?
This argument would be just as ridiculous saying "Sometimes I admire [some shitty brand] more than the supposed high quality Apple. [Some shitty brand] is a for-profit entity, they will take any cost cutting measure they can get (flimsy materials, shoddy electronics), and they make no bones about it. Apple is just faking making a good product because, you know, they're out to make a buck"
That sounds a bit shortsighted to me. I suppose it's ok to find Oracle's forthright greed refreshing, but... admirable? I find it awful.
I never thought of "don't be evil" as an end-all be-all promise to always do the right thing, no matter what. It's a guiding principle, one that has to take the back seat in the face of more practical considerations sometimes. We may not agree with all of those "sometimes," but that doesn't make those decisions automatically wrong. Google isn't perfect, but at least it seems like they try to have a social conscience.
Or maybe they've just done a good job of brainwashing me into thinking they do.
(does Oracle have any fans presently?)
No, Linux is not SysV based.
- Based on SysV standards - which GNU does for most commands, and which most Linux OSs did for initscripts.
- Based on Sys V code, which it obviously isn't.
The game still is survival of the fittest, but for this landscape, the fittest are good.
Without a doubt Google has the popular vote. I constantly hear people compare SUN to Google. Sun died, when? lol
Conceive a scenario where Google died tomorrow; now compare that to when Sun passed[it really did?]. It's not the same thing. "Google" is the most frequently used noun on HN, with almost 3 times the frequency of "Apple" [http://chegra.posterous.com/word-frequencies-in-front-page-h...]
Sun might be good for hackers, but Google takes it to a further extent and brought it to everybody.
Certain companies, besides their shareholders nobody cares if they live or die; nobody is super excited about their product. Nobody is saying they want to go work for them. The only reason why they exist is because they were best of the first movers.
It's common practice when a company gets big in order to innovate it purchases smaller companies. You don't innovate you die. All the big companies employ this strategy(really, when last Apple bought somebody-NeXT?). As an entrepreneur or a Hacker and being aware of the current situation, I think if you were going to sell your company you will stay away from Oracle; they will kill your culture. Also, given their business nature, they might even undercut you; it's more to your advantage to do business with someone who is good. And this is where I think Oracle has lost its advantage, it wouldn't be able to innovate.
Their Java move with Google is simple. IBM paid for Java, Google didn't. Oracle wants the cash.
I know a well-known privately-held extremely-high-end storage company who just canned its OpenSolaris/ZFS projects because they won't take the risk to be sued by NetApp.
ZFS as open-source is cold dead. Sad, but true.
http://www.google.com/patents?as_q&num=10&btnG=Googl...
They'll all expire eventually. ZFS (like) file systems will eventually be very succesful.
Yes, that's all in the "like". In 4 or 5 years, btrfs and others will mature and ZFS will look less exceptional. That, and ZFS has a major culprit which will become more annoying with time : it's not cluster-aware.
If there is any interest in OpenSolaris, someone will fork it and continue the development.