I worked at EMC at the time that Sun was aquired by Oracle. There were rumors that EMC was considering purchasing Sun to gain the final toe hold in the DC that they eventually filled by merging with Dell.
I am sure if EMC had bought Sun I'd still be there. Sun & EMC were both engineer led organisations in the old days, and the two cultures would have worked well together. (IMHO based on my experience working at EMC and with many many Sun alumni).
> Sun & EMC were both engineer led organisations in the old days
Isn't that exactly the problem? Server and storage HW were rapidly getting commodified. That's why the "vacuum cleaner salespeople" of the IT world (Dell, Compaq etc.) won, there's little value add in server HW engineering except at the bottom of the stack (Intel, NVIDIA, AMD, etc.).
It's also why HP, the actual engineer led scientific instrument manufacturer HP, left the computer industry and became Agilent leaving the vacuum cleaner sales people part behind. It's a crying shame, but you're right, PC manufacturers were and still are in a brutal race to the bottom.
For a while I though HP might have a chance on mobile, when they took on webOS. If they could have established a compelling platform stack with hardware, operating system and a decent dev environment they could have had a chance to grow their margins a bit. The problem is I suspect the PC side of HP just wasn't up to it. I had an HP digital camera around that time and it was a bulky, blocky uninspiring lump of a device.
Thats a little harsh towards Compaq. They inherited industry leadership with the 386 and their stewardship of the PC in that time was pretty damn good. Dell, the cow people etc...not so much.
Well, too bad IBM withdraw their offer, and if anyone thinks that they would care more about Solaris than Aix, IBM i or IBM z/OS, they are fooling themselves.
Databases they already had DB2. Informix did not had an happy life after aquisition, so most likely the love for MySQL wouldn't be that great.
And in what concerns Java, IBM has always had their own implementations, with extensions like JIT caches, AOT compilation, real time GC, special flavour of value types (ObjectLayout), so also here little value to get from Sun.
IBM would have continued Solaris development, maybe not made a major investment in it but would also not have destroyed it. They would have continued to develop it and ZFS in the open.
I simply don't see IBM doing what Oracle did and basically stopping OpenSource development.
And its not wishful thinking at all. Solaris development was happening in public it was well integrated with the community its not at all crazy to think that at least part of the development team would have continued working in the same way.
There were good products build on top of Solaris technology that IBM would have continue to develop and sell. Those products would not easily be built on top of AIX anytime soon.
IBM didn't need ANOTHER proprietary Unix, why would they go the Oracle route and make it proprietary again? What possible reason would IBM have to do that? The Openness of Solaris was what made it different form their other offerings.
IBM had also invested quite a bit of money in Linux and RedHat is literally part of IBM now. And apparently they have not acted the same as Oracle.
I understand why IBM didn't buy Sun that's not what we are arguing about.
Sun was selling products that could only work on top of Solaris. Had IBM bought Sun they would have continued to sell those products (as they were profitable) and likely continue at least some level of Solaris development to support those products.
They would have a had no reason what so ever to make that further development proprietary.
> tamed C into production
I have no idea what that even means. See has been successful in production on literally millions of computers and server around the world.
And if we are talking about usefulness to production Zones and DTrace are about 1000x more helpful then anything Oracle innovated.
> Something that naturally the Oracle haters don't care about.
Yeah people don't tend to care as much what happens in proprietary OS by shitty vendors.
Memory tagging isn't remotely new, LISP Maschines and many others had done it before. So by that count they didn't invent anything.
The code for it isn't even public so other could learn for it as Oracle made Solaris proprietary again.
> IBM invented timesharing and containers, no need for Solaris stuff.
I'm not sure what you are referring to by containers here. Please tell me where I can find a non-property Unix that allows me to use these amazing IBM containers.
You seem to place equal value on some proprietary system that people can't actually use, and those that made it available to everybody in Open Source.
Call me crazy but I value innovations more that are in the open and available.
> IBM withdraw their offer, because they didn't had any commercial reason to go forward with the deal, and they aren't a charity to start with.
I kinda wonder what IBM really is these days anyway.
They're a bit of a server manufacturer for some niche stuff. A bit of a CPU manufacturer for some niche uses. A bit of a consultancy operator. A bit of a software supplier. A bit of an operational outsourcer.
They're dabbling in everything yet don't shine at anything. No clear focus. If they didn't have their name and legacy I bet they would have been lost in the waves long ago.
They're an integrator and consultancy with a first-party mainframe platform, first-party data center play, first-party Linux development and support subsidiary, and the core developers for a lot of current-generation automation tooling.
What you get is 'Taligent'. A partnership between Apple/IBM in 1992.
> Taligent OS and CommonPoint mirrored the sprawling scope of IBM's complementary Workplace OS, in redundantly overlapping attempts to become the ultimate universal system to unify all of the world's computers and operating systems with a single microkernel.
Apparently it a combination of 'Talented' and 'Intelligent'. So what you get apparently is the worst aspect of both organization.
Its like if you would have Sun and Apple merge and you end up building a Sun GUI on top of a System 7 kernel.
Quite a bit of Sun software engineering ended up at Google so you're not wrong. Especially after Oracle shut down all Sun software development and laid everyone off in 2017 there was a huge wave of referral hiring.
Eric Schmidt was the CTO of Sun for a while and brought that culture to Google. I see a lot of the problems Sun had in Google and IMHO it's about to collapse under its own weight and lack of focus.
A lot of people at the time thought Google should have bought Sun. Google was heavily involved in Java at the time (with Dalvik especially), and got in a lot of trouble with Oracle when they bought Sun.
IBM has Eclipse. Not only is it an underhanded offense towards Sun, but IBM has shown that they are not willing to push the software to its best possible state.
Eclipse is nice but JetBrains' products and Visual Studio Code are nicer.
IBM is struggling with cloud computing. Adding SUN, they would have tried to turn it into a SPARK offer and burned much more resources.
Both SUN and IBM couldn't continue as before. How could IBM have used SUN to create something better?
At SUN's scale, there was no way forward for their OS or their chips. They were too late in open sourcing their OS, which could have created the market for their chips. Oracle did what SUN's management should have done years before.
In my opinion, Steve Jobs chose to build consumer products because that brings the scale to produce chips. It's in IBM's name to serve businesses, not consumers. The cultural change to create the foundation for SUN's hardware business is almost impossible to pull off.
I would like to argue that the world is a better place because Oracle is maintaining Java, something that SUN didn't fully achieve.
Instead of browsers and Javascript, we could have a jvm everywhere. We are building the full stack with technologies like react and virtual DOMs anyway. It's too late for that, but it's in Oracle's own interest to make Java shine.
But the world would have been a better place if IBM had bought Sun instead of Oracle.
Which bring me to this old joke:
Q: What do you get when you cross Apple and IBM?
A: IBM.