TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Why There Will Never Be Another RedHat: The Economics Of Open Source

59 pointsby kancheover 11 years ago

20 comments

hdevalenceover 11 years ago
It&#x27;s an interesting worldview where growing a company to an $11bn market cap counts as only a &#x27;lukewarm success&#x27;.<p>It&#x27;s not super clear why Red Hat&#x27;s market cap of &quot;only&quot; $11bn means that their business model doesn&#x27;t work. It would seem to me that it works pretty well -- to the tune of $11bn.
评论 #7234863 未加载
评论 #7237292 未加载
评论 #7234946 未加载
adwfover 11 years ago
So many shaky premises here that I barely know where to start.<p>1) Canonical. You could have made the exact same argument as this article 10 years ago and said that there wasn&#x27;t room for another RedHat. Canonical arrived and proved that yes, there is room for another profitable Linux vendor. I know I use Ubuntu server and am considering support contracts, so there is definitely a market.<p>2a) So Redhat doesn&#x27;t have the resources for development that Microsoft has and therefore will produce an inferior product? Have you seen Windows 8? Vista? Money does not guarantee success.<p>2b) Even without the same resources, the main Linux kernel has an enormous number of top-talent developers working <i>for free</i>. This is a serious economic advantage over traditional closed source companies.<p>3) Why on earth is the article comparing RedHat to Amazon? One is a Linux distribution vendor, the other is a global internet shopping mall.<p>I&#x27;m sure others can provide more reasons. Or even counterpoints; feel free to rebut ;)
评论 #7235518 未加载
评论 #7235313 未加载
评论 #7235214 未加载
评论 #7235047 未加载
arca_voragoover 11 years ago
Please, don&#x27;t start pushing SaaS into FOSS under the guise of profitability. Sure, I&#x27;ll admit that the support model (RedHat) has it&#x27;s issues and probably keeps profits lower than competing proprietary solutions, but lets not forget the third model: up front sales.<p>Here&#x27;s a problem I run into frequently: I have a problem, but as a look for a software solution to that problem, I find 3 potential solutions. Of those solutions, one is a random sourceforge mish-mash that hasn&#x27;t been maintained in years, and the other two are proprietary closed source solutions (which I try to avoid whenever possible. Why should I trust some random small company I&#x27;ve never dealt with to not be malicious or incompetent?)<p>What I really want is a good, up to date software that I can pay for, but is still open source. I think too many people have forgotten that open source doesn&#x27;t mean that you have to release the code in public for all to see on github et al, it just simply means that you have to provide the source.<p>So unless you are a customer, you don&#x27;t get access to the source (unless the customer then shares it... this is probably one of the few areas I think that needs more debate in the GPL vs MIT war.)<p>With all of the revelations of the NSA, I think any business that deals with sensitive information in any way has a responsibility to 1) move away from SaaS and cloud computing, and 2) move away from proprietary whenever it is possible to do so without negatively impacting business operations.<p>We need to reclaim control of our data, as users personally and in business, and the only way we reliably do that is with decentralized, locally controlled FOSS.
wmfover 11 years ago
I have a small disagreement with the market cap&#x2F;revenue argument: &quot;its key point of failure is that the business model simply does not enable adequate funding of ongoing investments.&quot; Red Hat never intended to become as big as MS (they&#x27;ve repeatedly claimed their goal was to actually shrink the OS market by selling Linux cheaper) because they don&#x27;t have to fund the entire development of Linux themselves.
评论 #7235515 未加载
davidwover 11 years ago
The economics of open source software is fascinating.<p>I think the best model for open source is to be the &quot;base of the pyramid&quot;, with companies adding a little bit of proprietary stuff that differentiates them at the top, and collaborating on the infrastructure below. Stuff like Ruby on Rails works this way, with Basecamp being the point of the pyramid, and Rails, Ruby, Linux, Mysql etc... being all the stuff it&#x27;s built on.
评论 #7235209 未加载
jlawerover 11 years ago
The redhat model is really an outgrowth of the open source consulting model. If you look at smaller companies there are hundreds of projects that have spawned companies doing consulting, custom development and professional services.<p>Red Hat is unique mostly because they have been able to grow to this massive size. Part of this was initial capitalisation, part of it was right time, right place and mostly the fact that they have been able to add value to end customers more then their competitors.<p>XenSource had the problem of many other providers taking their output and repackaging it into a better product for customers (Amazon takes Xen, makes AWS. Other companies build their own virt products around it. If Xensource was the best virtualization platform and people had flocked for it, they would have been able to build a solid revenue from enterprise customers who required the guarantee of support for their mission critical workloads. Unfortunately there are lots of virtualization platforms around and I never found anything compelling in XenSource&#x27;s Offering. VMware &amp; HyperV have all the 3rd party support, oVirt &#x2F; RHEV &#x2F; OpenStack fill the cheap end.<p>I accept that with what the market wants Open Source companies of the size of Red Hat are rare, mostly because investors want larger returns, however this doesn&#x27;t mean a open source company can&#x27;t be profitable and grow to this size, it will just be significantly slower. Companies in the space typically will have a slow but solid growth pattern when firing on all cylinders, Not the &#x27;hockey stick&#x27; growth curve many investors are looking for.
anon4over 11 years ago
I&#x27;ve been wondering for a while if a company can work on purely open-source software and profit from the software itself, rather than only from the stuff that happens around it. The idea that I&#x27;ve been thinking over is a subscription for voting on features. That is, a user of the software will pay some monthly amount and get some number of votes for directing what new features will be implemented. Making sure that votes work properly and oft-requested features by people with low voting power still get implemented timely is one problem. Another is for companies that just want one bug fixed or feature added and aren&#x27;t interested in long-term contribution. They could be easily outvoted by long-term contributors.<p>Has something like this ever been tried and how did it work or fail?
评论 #7235667 未加载
cwyersover 11 years ago
What a ridiculous comparison of apples and oranges and peaches and pears and plums and bananas. Amazon isn&#x27;t even remotely in the same business as Red Hat. Microsoft is in the business of providing enterprise operating systems, but that&#x27;s far from it&#x27;s only line of business. Using Oracle as an example of how &quot;selling support for open source software&quot; isn&#x27;t a repeatable business model ignores how Oracle is copying Red Hat&#x27;s model, and in fact some of Red Hat&#x27;s source code.<p>Is it true that Red Hat doesn&#x27;t have the money to invest in improving RHEL? I don&#x27;t know (although I suspect it&#x27;s not true), but the article spends zero time on anything that would demonstrate that.
jfasiover 11 years ago
What prevents RedHat itself from transitioning to the open source platform model this article touts? They&#x27;ve put out OpenShift, which is a Heroku-style deployment clone, so they&#x27;re at least making steps in this direction.
Uhhrrrover 11 years ago
There are other RedHats already, SUSE and Canonical among them. So the article is flawed halfway through the title.
dredmorbiusover 11 years ago
You could apply logic similarly to why there will never be another Microsoft.<p>If you look at it, Microsoft was a tremendous outlier. Not only was it the biggest licensed software giant, it was the <i>only</i> licensed software giant. By profit margin, until Apple and Google came along, nobody could compare. Cranking out CALs is literally printing money. And by revenues, the only companies which <i>could</i> compare were principally large integrated service providers: Oracle, IBM, SAP, PeopleSoft and Big N consulting firms&#x27; software divisions, producing custom software solutions selling brains by the bucketful (a notoriously difficult business to scale).<p>As wmf notes here, Red Hat&#x27;s stated aim was to take the money out of the OS market (and put it in users&#x27; pockets), a task with which they were eminently successful -- they de-monetized Sun right out of existence, along with most of the proprietary UNIX market and much of Microsoft&#x27;s server side.<p>The real value of free software is in enablement, which is why I see Google, Yahoo, Amazon, Facebook, and other companies built on the back of free software as its true legacy.<p>Of the companies Levine compares RH to, Microsoft is stumbling badly and Oracle has pretty much proven itself the place free software projects go to fork. VMWare is based on services to the free software (and proprietary) world, Amazon I&#x27;ve already defined as being build <i>on top of</i> free software.<p>This article is sorely nearsighted.
评论 #7236861 未加载
mindcrimeover 11 years ago
Oh golly, gee whizz, I guess we should just close our doors and shut down then. Peter Levine thinks you can&#x27;t make money as an Open Source company... well, that sums it all up.<p>Wait a minute.<p>Fuck that.<p>I know Peter Levine is smart and well-respected and all that, but this is so much bullshit. For starters, all he really said, for using all those words, is &quot;it&#x27;s hard to build a business to the size of Oracle or Amazon or Microsoft, around a core of Open Source&quot;.<p>No. Fucking. Shit.<p>Really?<p>It&#x27;s hard to build a business like Microsoft, Amazon or Oracle??? You&#x27;re kidding me, right? I mean, it&#x27;s <i>not</i> like any of those companies are atypical outliers in any way, right?<p>So, if we just take all our source, close it, and move to a proprietary business model, then we should have no problem building &quot;the next Microsoft&quot; right?<p>Also, am I wrong in thinking that Amazon is hardly even in the same business as Red Hat? What are they even doing in this comparison?<p>I dunno, color me biased (I am) but isn&#x27;t Peter <i>really</i> just arguing that an &quot;open source&quot; company isn&#x27;t going to generate the returns necessary, in the required timeframe, to justify investment by Andreesen-Horowitz? Because honestly, that&#x27;s all I&#x27;m hearing. Nobody says you have to become &quot;the next Microsoft&quot; to be successful... well, nobody except Peter Levine, I guess.<p>Meanwhile, SugarCRM, Alfresco, Red Hat, Cloudera, BonitaSoft, JasperSoft, Pentaho and a whole laundry list of companies are making money &quot;selling open source&quot;. Are any of them going to IPO? I don&#x27;t know, but from where I&#x27;m sitting, that isn&#x27;t the point.<p>All of that said, where I wholeheartedly agree with Peter, is the bit about adding a SaaS element with the underlying technology as a platform. At Fogbeam, I expect we&#x27;ll have a traditional &quot;support subscription&quot; (ala Red Hat) model going for a long-time to come, but we are definitely starting to move in the direction of building purpose-specific &#x2F; vertical aligned solution on top of our base stack, and delivering those as SaaS offerings. Personally, I see those as complementary strategies, and not mutually exclusive.
cdoohover 11 years ago
I don&#x27;t really see why this may be a bad thing, the open source community continues to grow just fine and code contributions and stuff still happen. When I first read the title thought he&#x27;d be talking about how it&#x27;s impossible to make money off opesource but that wasn&#x27;t the case
pessimizerover 11 years ago
Who cares about revenues? Of course the revenue is going to be higher at companies that sell software and hardware rather than support it. Costs are going to be higher too, for a company that has to write all of their software without help.
na85over 11 years ago
I started reading this article waiting jadedly for the inevitable call to move to &quot;X as a service!!!!1oneone&quot;.<p>Sure enough, the article did not disappoint.<p>XAAS pretty much requires a cloud-based system, and public trust in the cloud is rapidly diminishing.
mindcrimeover 11 years ago
I just realized that there is another point that Peter totally skips over in this analysis... the value of community good-will, and the effect that has on the economics. For example, I&#x27;d bet money that a far larger percentage of Red Hat&#x27;s customer base fundamentally <i>like</i> Red Hat, than the corresponding percentage of customers who <i>like</i> Oracle or Microsoft. Maybe I&#x27;m wrong, but I suspect that RH have lower customer acquisition costs and don&#x27;t have to spend as much on advertising, since their product itself (in open-source form) <i>is</i> advertising for the company.
brudgersover 11 years ago
Joel Spolsky talked about building a free platform and living off the exhaust fumes here:<p><a href="https://news.ycombinator.com/item?id=6462430" rel="nofollow">https:&#x2F;&#x2F;news.ycombinator.com&#x2F;item?id=6462430</a><p>That&#x27;s really the RedHat model. They just happened to get in early in an area that scaled well.
soundsover 11 years ago
I agree with Levine&#x27;s conclusions, but the reason Red Hat is successful in spite of &quot;selling a slightly better road&quot; is simple. (The car analogy really helps this.)<p>Unlike the other open source efforts Levine lists, Red Hat sells Free Software. Now, hang on just a second and hear this out: it&#x27;s not that Red Hat is _evangelizing_ like the FSF does.<p>Red Hat has simply capitalized on the model that GPL Linux is better than _any_ proprietary server OS. I know I&#x27;ve never seen anything come close to it. Feel free to point out a server OS that competes successfully with Linux – gratis!<p>Red Hat can&#x27;t charge for the razors but they make up for it by charging for blades: stuff your company _shouldn&#x27;t_ be doing in-house. You can outsource it all to Red Hat: installation, maintenance, QA, everything. They&#x27;re basically an early freemium model.<p>Levine is correct that the market potential has not exploded for Red Hat (yet) but he concludes that nobody else should try what they&#x27;re doing. I think he doesn&#x27;t understand exactly what Red Hat is doing.<p>Due to the unique pricing structure caused by Free Software, a &quot;differentiated service on top of the platform&quot; model would fail. For Red Hat, that would be something like selling proprietary system management software on top of Linux, and assuming it was sufficient.<p>If Red Hat were evangelizing Free Software they might be content with a sub-par product &quot;because freedom!&quot; Take laptops, for example: System76, ZaReason, and laclinux.com remain small. Their product does not represent a tier 1 product, even though they make a differentiated laptop on top of the same base of laptop ODMs.<p>Levine correctly concludes that Red Hat will not grow as rapidly as a Google or Amazon. It&#x27;s not an apples-to-apples comparison. Instead, compare Red Hat to Microsoft, CA, or Oracle who compete directly in the &quot;razors &amp; blades&quot; market of enterprise IT support. (Those companies do have other markets as well.)
评论 #7235323 未加载
vainover 11 years ago
The article isn&#x27;t very well thought through. mongodb is doing very well just selling support
tonydivover 11 years ago
What about Cloudera?