Hacker News new | past | comments | ask | show | jobs | submit login
Before sale to IBM for billions, Red Hat started in cofounder's closet (cnbc.com)
121 points by goldminer88 5 months ago | hide | past | web | favorite | 37 comments

I have a fond memory of the original redhat distros involving a closet. Sometime in the mid 90's, I was installing RedHat on an old (even for the time) 486/33 server that we had managed to acquire with the understanding that it be relocated the next day to a closet in another building across campus. Since the machine was actually on a desk, we thought we should do the install before having to move it to a cramped space. The installation ended up taking like 18 hours on that box, and it came time to move before it was finished. So we unplugged the UPS and had a bunch of people walk the thing across campus along with its tiny monochrome monitor still showing the RedHat installer filling up progress bars over and over. It prompted a lot of questions and strange looks. We were worried that the batteries were going to give out, but we made it. The machine carried on dutifully for many years longer than it should have. The Redhat distro along with rpm really made Linux a workable and supportable platform, and they deserve a lot of credit for their efforts.

Its actually APT (apt-get) which made Linux a workable platform because before that we had to manually hunt for package updates (such as RPMs). That was horrific. It is the sole reason I ever went to try (and switch to) Debian, coming from RPM-based distributions.

Yup. I've been using linux/unix since before package managers were common and debian was like a breath of fresh air. Even with RPMs it was generally always just easier to build stuff from source than to deal with RedHat's funky package system. And APT? APT was easy mode, it worked like magic. If you needed something you just installed it and the OS picked up the latest package and all the dependencies and installed all of it for you. It seems trivial today but back when always connected broadband internet was not the norm it was transformative. It took years and years for RedHat to catch up. I still have a sore spot about RedHat because of those earlier experiences.

Things are so much easier now in so many ways. For one it's a lot less common to have a computer that can't access the internet because it doesn't have the right drivers already. For another it's a lot less common to only have one computer (including smartphones) and thus only one way to download stuff. Computers are also way faster so doing anything that requires installing or compiling stuff isn't as big of a time sink.

My first exposure to RedHat and linux was playing with a Toshiba Magnia sg20 internet appliance. I found out I could telnet into it and get root access where it was running RedHad 6. I messed around with it a lot, just learning things and trying to get as many cool console apps on there as I could find. The pain of doing manual package dependency is why I jumped over to Debian (this would have been around 2003 or 2004). Though I do consider the experience I gained in doing manual dependency resolution valuable.

rpmfind.net !

> "The $34 billion success has nothing to do with me. I will take some credit for spotting the initial opportunity. The actual execution of that opportunity, the credit goes to much smarter guys than me."

I respect Young for such an honest attribution.

"Young now calls his wife "probably the single-most important contributor" to the initial success of Red Hat. "Without my wife Nancy's sterling credit rating, I wouldn't been able to raise the money that got us to profitability," he says."

I can relate to this.

totally appreciate his humility but also think he can take a little more credit.

i'd (briefly) considered creating a commercial linux company way back when, but didn't have the wherewithal or gumption to do so. a key ingredient is being able to get started, attract and accept help, and then have enough sense to give others room to work.

He did leave the company 13 years ago as far as I can tell

That might be, but by the late 90s my first few stabs at Linux were using Red Hat because it was far enough along that books were published and available at my local book store (I did also try Yellow Dog, a Red Hat variant, when repurposing an old PowerPC). By the early 2000s I saw box sets in Best Buy. In my industry, most of the large companies moved from IRIX to Red Hat and have been using them since. Most of the others moved to a different free distro, then to CentOS.

Sure, and it seems like he's happy to take credit for that era, but most of Red Hats value came from the time after Young left, and I don't really think it's modesty for him to not take credit for that

Red Hat was initially financed by opening 8 different credit cards and accruing $50,000 in credit cart debt before they began bring profitable. I never knew this before but really drives home how hard/risky it was to bootstrap your own business back then.

Loads of people do that now. I bet many restaurants/food trucks etc got started with credit card debt.

why the "back then" modifier? it's still hard/risky now.

1993-1994 was before the first flush of "dotcom 1.0" era VC money, which showed up in the 1996-2000 era before the big crash.

Existence of VC money is irrelevant to 'bootstrapping'. Taking VC money is sort of the opposite of bootstrapping.

Existence of easy seed and pre-series-A funding nowadays means people don't even try to bootstrap with 50k in personal 20% APR debt. Similarly in 1998, 1999, a lot of wild dreamers with harebrained ideas got VC money pretty easily after a few PowerPoints and some bullshitting.

Google also used a bunch of credit cards initially from what I understand.

I'd rather have 20% APR, than give away 20% equity. Why? Because when the debt is paid to the credit card company they go away. The VCs will always want more and more, pushing your company faster than it should go.

Depends on risk profile.

30 with 3 kids and no savings? Yes please, give away some of my profits so VCs can derisk the startup.

20 and living with parents? Bootstrap away, if I fail I will be fine.

IIRC the google guys had VC crawling down there back and had to tell them to back off.

Based on personal experience, I doubt $50k in credit card debt was what was planned for and probably came out of necessity to finish the thing.

Fund-raising takes a lot of mental space on its own which comes at a high premium when you're focused on finishing a product.

Even with $50k at 20%, we're talking about $833 a month in interest, that's not that much to someone who would usually make $100k+ at a job if they fail.

Opportunity cost is the real liability.

20% of $50K is $10K/year which is like 15% of the take-home pay on a $100K salary, and all you did was pay interest. That's not something to sneeze at.

Interestingly, the kernel of IBM (Bundy Manufacturing) started in 1889 with $150K. https://history-computer.com/People/BundyBio.html

I wonder how common that funding level was in the late 19th Century.

To put things into perspective $150k 1889 would now be worth about $4.1mil [0]

[0] http://www.in2013dollars.com/1889-dollars-in-2018?amount=150...

Most of the major success stories you're familiar with from that era, involved entrepreneurs that had rolling levels of success. They built up reputations first, which enabled them to raise large amounts of capital as needed to fund ever larger ventures. Elon Musk's latest successes are of that mold (which makes sense, when you consider that his latest businesses share more in common with old industrial ventures than dotcoms; Zip2 and PayPal gave venture capitalists and other partners the confidence to back him on Tesla and SpaceX).

Rockefeller did that for example before creating Standard Oil. He built up a sterling reputation first working as a bookkeeper, then as a partner at a trading firm. He got to know the owners of capital in doing so. At times during the early years he required extraordinary sums of capital, his reputation enabled him to raise it in an evening by riding around town and meeting personally with bankers. He did that once to buy out a large shareholder who had become conflictory; and other times to acquire businesses quickly.

Henry Ford went through multiple iterations of messing around with commercializing his creations (including the Henry Ford Company), before finally founding The Ford Motor Company. After his very early experiments with vehicles, his later efforts all required the support of financiers and lots of capital. Over time he had demonstrated a knack for pulling complex engineering efforts together and attracting talented people.

Edison always had capital backers for each of his major commercialization efforts. He operated his own lab (at times funded by outside capital) and then typically made arrangements for each set of inventions with financiers to take them from prototype to commercialization and distribution. Nikola Tesla had some similar backing arrangements at times, although his history in business was even more haphazard.

Before George Westinghouse founded Westinghouse Electric (which is what made him particularly famous in the era), he started with The Westinghouse Air Brake Company, which he founded at 23 years old. And even before that, he started by creating useful industrial inventions, receiving the first patent for a rotary steam engine at 19 and a bunch of smaller railroad-related inventions, before striking commercial gold with the air brake.

Interesting. Has anyone ever written a book about founders across the ages? I remember that the US was always a bit Patent crazy and wonder how that translated into getting capital. Bundy invented the time clock but seems to have been a successful marketeer for his earlier clocks.

The Wright Brothers blew a lot of energy defending their patents and not competing in the market.

The balance between hard skills, showmanship and ruthlessness would be an interesting study.

> While at CMU, Marc Ewing was known to wear a red hat as he walked between classes. Because of his computer expertise, people would ask for help from the "man in the red hat". Ewing and co-founder Bob Young named their initial software after the hat, and the name Red Hat stuck.


Still.. No one talks about him any longer, and the Wikipedia article is so short that we cannot know what else he did after leaving Red Hat..

Apparently he left the tech industry, co-founded the Alpinist[1] (he has sold it since), and started pursuing sailing on a competitive level.

He has also a sparse Github profile: https://github.com/marcewing

[1] http://www.alpinist.com/p/magazine/AboutUs

What the...? I had no idea about this. I am a subscriber to Alpinist. In fact, I renewed my subscription not too long ago via one of my laptops, which is running Fedora. Thanks for a couple of great things, Marc.

As a Red Hat user and Alpinist reader, thanks for sharing this. I never knew the connection.

That's it, I'm moving my startup under my stairs.

Yer a wizard Harry!

props. humble guy. credit to the rpm guys... binary builds from clean untouched tar files sealed the deal for me back then. i still miss the box sets.

- The lower-level the category the more customers want support on it.

- The more you can commoditize the layers below you, the more the market likes it. Red Hat does this for servers.

- The lower-level the category the more the market actually “wants” it standardized in order to minimize entropy. This is why low-level infrastructure categories become natural monopolies or oligopolies.

source: https://kellblog.com/2014/11/18/it-aint-easy-making-money-in...

The article portrays this as a success story, but was making “billions” the point or goal of Red Hat? If it wasn’t, then they have now failed. Because that’s all you get when you “exit” like this – you get money.

Didn't someone in SV say "Software is eating the world"?


I suppose 34B is a good run. I mean its not $110B, but its not Biotech so.

Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | Legal | Apply to YC | Contact