An old (2010) story from former Sun employee Jeremy Allison.
David Miller wrote (at the end of a long email explaining how Sparc Linux used cache optimizations to beat Solaris on performance):
“One final note. When you have to deal with SunSOFT to report a bug, how “important” do you have (ie. Fortune 500?) to be and how big of a customer do you have to be (multi million dollar purchases?) to get direct access to Sun’s Engineers at Sun Quentin? With Linux, all you have to do is send me or one of the other SparcLinux hackers an email and we will attend to your bug in due time. We have too much pride in our system to ignore you and not fix the bug.”
To which Bryan Cantrill replied with this amazing retort:
“Have you ever kissed a girl?”
Talk about missing the point and underestimating the competition.
The article offers an interesting look at why Sun eventually failed. I stumbled upon this story because Rob Landley mentioned the girl comment in his email about BSD/SysV.
Simply outstanding piece of history. We used to believe in Sun/Solaris back in a day (around 2002-2005) on our Java application servers but even tiny Thinkpad laptops could outperform the big fancy Sun boxes. However this article sheds a light into why this happened.
I remember being astounded with the speed of the first SUN machine I used (SPARCstation IPX, 1993) but that was coming from a background of out of date high school 386sx PCs.
Then upon graduation (1998) I was again impressed with the Ultra10 workstation is got in my first job, The performance was again fantastic, despite the pathetic IDE hard disk system those machines used.
Still, SUN was not always about absolute single thread speed. I bet those tiny thinkpads couldn’t scale up to 64 CPUs and 10s of GB of RAM, which Solaris machines of the time could easily do (with the right $$).
I guess SUN got by on the ability to keep scaling up, if you were willing to pay for it, before the advent of massive server farms and distributed computing. Alas, distributed application stacks put paid to that, and the writing was on the wall.
The SunLinux guys were smarter/more dedicated/more cleaver?
That Sun underestimated Linux?
That the SunLinux guys effectively killed themselves off by killing off Sun?
That anything outside technical development is pointless?
That the Sun guy was a jerk?
I’m not sure I agree with any of those inferences.
The sun guy isn’t a jerk at all, he’s a nice guy. But the point is that he thought Miller was a jerk, and that’s kinda true (albeit technically a marvel.)
How do you turn it around so that the person on the receiving end of the dismissive insult becomes the jerk?
Dang. That’s a good point. From a certain point, Miller really is being a jerk in that.
I interpreted Cantrill’s comment as the mid-90s equivalent of “TL;DR” or “cool story, bro.”
And who most often say one of those two things?
People who quote nearly 2,000 words of text just to add a one-liner?
To me it seems the sun guy actually confirms he is not serious about the topic. This Miller dude is aiming for an intelligent discussion among professionals about their business. The reply is disrespectful and proves the sun guy is not interested in the topic.
So, yes I think he is a jerk. It’s an attitude which probably attributed to suns ending and the rise of the Linux machines.
Probably not.
Linux was technically on par, not all that better. However, it had a much better price tag. Free.
Edited 2015-01-29 20:20 UTC
klagermkii,
I think it was intended as a joke more than a serious insult or rebuttal. Something you might say to your pals when your kidding around. Adding “LOL” might have helped because the internet doesn’t convey emotion well.
Indeed context makes a big difference, but I don’t think one gets the impression that David and Bryan were ever pals and the whole thread feels like antagonistic rivalry. I don’t think in that context one can assume it’s just banter between friends.
Also remember that in the era this was written (1996) the attitude of pretty much all major vendors towards Linux was dismissive condescension, so comments coming from a Sun employee would be taken in that kind of light.
krakal,
Yea, many great technologies came out of Sun, I think they deserved a lot of respect for their contributions. Despite Sun’s tech having certain advantages, taking crippling blows from the Microsoft monopoly, and then on top of that having a unix competitor that offered free and mostly compatible alternatives proved to be too much.
Have you ever used Solaris? I’m guessing not. And certainly not circa 1996. Linux was not (and in many way is not) technically on par with Solaris.
I guess you also must have not used solaris much back then either. In 1996 using the term “slowlaris” was not seen as a diss against the system, but as a statement of fact.
Solaris was a mess until the 2.6 release really. Sure, linux back then still lacked some of the facilities Solaris was offering. However, the fact that a hobby project was able to match or better the performance of solaris, on its own native platform and at an infinitesimal fraction of the cost, was quite the eye opener.
Have you ever used Solaris? I’m guessing not. And certainly not circa 1996. Linux was not (and in many way is not) technically on par with Solaris. [/q]
I used SunOS and Solaris since the very beginning as a telco engineer. I may be old, but your guess is wrong.
Or, put differently, Sun had the disadvantage of being ridiculously overpriced.
Just like Oracle these days
In 1996?! Are you serious?
Yeah maybe to play at home, what big boys were using at big companies.
In 1994 it barely supported CD-ROM drives, by 1996 setting up X still required lots of work playing around with modconfig assuming one was lucky with the graphics card and monitor.
3D acceleration wasn’t even a thing on GNU/Linux.
By 1999 our university was still using DG/UX and Aix for tasks where being a real UNIX still mattered.
No, Miller can be seen as bragging about how leet the sunlinux team is. The sun guy didn’t write the calling convention for the OS, probably Bill Joy did. Who is pretty dang awesome of a guy. He probably did it to maintain compatibility with some important apps/ systems of the day, which by the mid 90s was probably irrelevant. Its like a rich guy asking why the poor guy is so stupid to choose to be poor. He can’t help it, there are forces beyond his control that keep him poor.
It’s 1996. That isn’t poor little Sun being picked on, that’s around the era of “we’re the dot in dot-com” Sun.
I don’t think you can read Bryan’s reply as if he’s the poor kid who’s been offended by the rich kid picking on him. No he’s the rich kid mocking the poor kid because his dad is the biggest name in servers, and his unassailable wealth means all the “stupid technical reasons” aren’t even worth discussing.
Dude, there is nothing Bryan (The sun guy) can do from a technical perspective to improve solaris/sun os. The calling convention can not be changed. Its not a debate over what is the best way to do something. Its obvious that the linux way was better. And you have the Linux guy bragging about how much better it is. The sun guy doesn’t respond on a technical level because there is nothing to say. A yugo *is* an inferior car to a porshe.
That’s ridiculous, making a strong technical argument is not being a bully. If you can’t improve the state of the operating system you’re defending and you’re finding it difficult to rebut specific points, you don’t just get to instead make dismissive personal insults.
Every day forums are full of people who want to defend their chosen “team” of Android, iOS, or Windows Phone and if they cannot “win” technically and decide to personally lash out, then they’re being jerks. Just like this guy.
And this Bryan guy is being paid a decent salary by Sun, so the least he could do is not be a jerk and reflect the jerkiness onto his employer.
He had many other options available to him. He could have not replied. He could have emphasized Sun’s strengths instead: “1000s of companies trust and use Sun”, “In the commercial market we emphasize reliability and consistency over pure performance”, “We have a global presence and can send people onsite to assist”. Instead he just acted like a petty jerk.
Careful, I wasn’t saying he was justified in doing anything. But I understand why he may have felt offended at the email. No one gets to harm anyone. What you don’t understand is badgering someone over a completely uncontestable point can be offensive as well.
Well calling him a jerk or not rests on whether he was justified. If he was just someone fighting back from being harassed, then he’s probably justified, and that would make him not a jerk.
Again, being offended because someone is denigrating an inanimate product you support is understandable in the sense that people incorrectly take it personally when you highlight flaws in their favourite brand of car, mobile phone, sports team, operating system or game console.
Bryan hadn’t been involved in the thread before, no one had even been talking to him, he just injected himself into the conversation with his dismissive one-liner. There hadn’t been any badgering of him.
If someone somewhere on the Internet writes something very negative about a product you support to someone else, is that now called badgering?
Not in my book. My ethics are different than yours.
It’s seems pretty hard to disagree with the last one.
Sun was still more-or-less successful when bought (and destroyed) by Oracle. Some people thought Sun WOULD fail, but it never got the chance. And yes, people at Sun were cocky! But the business was good enough that Larry paid the price to destroy a competitor. Sun didn’t fail.
Larry didn’t buy Sun to destroy a competitor, he bought Sun for Java plain and simple. With most of Oracle’s application stack running on Java he wanted/needed that critical piece of infrastructure.
I agree to disagree!
I don’t think the word “successful” means what you want it to mean. Sun was hemorrhaging cash big time.
There are successful businesses trading publicly which lose money! Sun was in trouble but had not failed yet!
Sorry, but the only things I can extract from your reply is that either your cognitive dissonance is very strong in this matter, or you do not understand what the term “successful” actually means in the context of business.
Edited 2015-01-31 20:10 UTC
Maybe you just have don’t have enough experience to know better! Losing money does not imply failure. Selling out does not imply failure. Some of the biggest names you’ve heard of lose money.
One must concede that Sun’s biggest stakeholders may have believed failure was unavoidable, and that’s why they sold out first. Either way, Sun was gone before it could fail.
Or maybe you have had a lot of experience, and yet you have not managed to grasp the basics: A public company that makes profit, or is in a clear position to provide investment return = successful, a business that loses money and can’t provide a clear path for growth in investment return = not successful.
Guess to which group SUN belonged at the end of its run?
That may be so, but in the case of SUN it did.
Too bad that in the case of SUN it did.
So? I never implied that SUN was the only corporation failing.
Again, your cognitive dissonance is working overtime to make failure not be failure. If sun’s biggest stock holders decided it was failing at providing returns to their investment, then by definition sun was being a failure at it’s very reason to exist: to provide value to its stock holder.
Seriously, it is not rocket science; SUN failed, end of story. Had Oracle not stepped it, or another player with cash for that matter, they would have had to close doors. They simply did not have any clear return to profitability path on their own. It’s a pity, I really liked some of their technologies and I worked extensively with people in that organization (I even had an internship there).