I can't find the blog post, but some blog wrote a while ago why VCs invest in companies like Github. TL;DR -- basically it provides infrastructure for other startups.
The business itself may not be a great business due to the amount of cost it takes to run it -- but it's necessary for the running of other ventures.
Except as pointed out in the article, the massive increase in spending was fairly frivolous. They spent crazy money on a new huge office, threw big parties, and sent their employees traveling everywhere. They also doubled their headcount, when I think we can all agree that they haven't had a surge in volume or features that merits that many new employees.
There's no reason that GitHub couldn't be run profitably if they weren't just out there burning VC money as quickly as possible.
It may have just been a coincidence. I believe they brought major features online two or three weeks after the HN-hatefest against them started, hardly enough time.
I'd also like to point out that Github was probably the most important change in OSS software by a wide margin. People easily forget comfort they've grown accustomed to, but it's worth to take a trip into the past from time to time: https://sourceforge.net/projects/avogadro/?source=frontpage&...
(and that's today's sourceforge – they didn't do much, but 10 years ago it was definitely even worse)
They built a replica of the Oval Office as their lobby. I don't think it lasted more than a year. It's all gone now, and not a spec of the original Oval Office is there anymore. It's an open seating area for coffee...
New features doesn't generally scale linearly with employee count (due to required communication paths, etc.) That said, I actually do think there's been an uptick in useful new features from GitHub this year. We use Phabricator at Khan Academy, and many aspects of Phab that were superior last year when I joined KA are now in line with GitHub's offering.
Phabricator definitely has a superior sense of humor. (If anyone reading this hasn't checked out their website, I highly recommend it)
I was, of course, not trying to say that Phab was regressing but rather that GitHub had picked up the pace. There have absolutely been good improvements to Phab this year too!
In the case of Phabricator, I wonder if that sense of humor works against them. The first time I followed a link to their site and read the description[1], I was convinced that it was a joke poking fun at how ridiculous overwrought do-everything enterprise systems tend to be.
Are you saying that VCs are investing because they see companies like GitHub as being public goods? That's contrary to normal investing principles, unless there's some reason to think that a VC investing will result in their portfolio companies having superior access.
The point is more or less, if VC invests (or plans to) in 20 software companies, investing additionally in something like GitHub or npm makes it more likely that the other 20 companies will succeed. They don't invest just in GitHub from the goodness of their hearts.
I know nothing about investing but I feel somewhat skeptical about these arguments. How many startups actually failed because they were using a poor package manager or source control host?
This sounds more like a spin-off of the old investment adage: "Buy stock in the products you actually use." If you spin that argument slightly as a proxy purchase, it makes sense that VCs would be investing in the products that their other startups are using heavily.
As an super-small-time investor who only owns stock in companies whose products that I use and enjoy, I can't knock them for it. For me, the logic is that whenever I tire of using something or no longer find it valuable, presumably I'll have early insight to sell the stock before the rest of the world catches on. I don't know if that same logic applies to proxy buying, but I suppose if you're intimate enough with your companies to know if they're abandoning Github for something else, as I don't know if pulling venture capital is as easy as selling the stock.
> as I don't know if pulling venture capital is as easy as selling the stock.
There's generally minimal liquidity. During a round, an existing investor may have the opportunity to sell some shares to new/other investors, but if she knows something that's not coming out during diligence, there's definitely something fishy going on. If the round is shaping up to be a major up-round, maybe an early investor wants to lock in a good return, but that's beside the point here. And then of course if things really aren't going well for the company, you're looking at the bad kind of liquidity event--a liquidation.
So if a VC wants to pull out based on a negative hunch, it's probably either impossible or the signal itself will doom the company if it wasn't already doomed.
A startup that can't iterate quickly enough will die by a thousand cuts, and it will be hard to say exactly what killed it. I'd certainly think using github can extend your runway by 5-10% (just in terms of how much time it saves we as a developer), and in a certain proportion of cases that will be the difference between success and failure.
To add, if Github goes into a death spiral they'll likely take a few of those 20 companies with them. In the same way if AWS goes down, so does half the Valley.
I'm actually curious how true is this. Most of the companies I talked to / read about use a very small set of services which are popular everywhere (compute + storage) or services which can be relatively easily distributed across other providers (for example outbound emails, event notifications, etc.) Projects that really use a lot of the AWS and are really tied up in that environment are much less common.
Is that a skewed view? Are there many big companies that would disappear without being able to move to a different platform in under 2 weeks?
I am hugely skeptical that this ever happens. Do you have any evidence?
I believe no VC is going to invest $50 million dollars purely in hope of vaguely helping the rest of their portfolio. While at the same time vaguely helping every other startup in existence, thus negating any advantage that might accrue to their own companies.
think people in the industry are just desperate to try and find justification and sanity where there really is none because it makes them feel the whole system is more stable than it truly is.
The reality is there is no reason for GitHub to have this much funding and spend so frivolously. This should be a warning.
Actually... without any proofs whatsover, who knows who is really glancing over private repos parked at GitHub. I bet there are many projects with multi-million dollar source code (aka secret sauce) that many would love to look into.
> basically it provides infrastructure for other startups
The good news for them, is they can become much more. GitLab learned earlier on, the value of selling to Enterprise as opposed to startups. They (GitHub) really should have gone on a hiring blitz a few years ago, to find people who understood Enterprise.
GitHub, way over estimated the value of "social programming", when it comes to Enterprise. I would say 80% of programmers in Enterprise, are not passionate about programming and have no interest, in the "social" value, that is offered by GitHub Enterprise. The vast majority of Enterprise programmers see it as a job, and really don't care about what others are working on, unless it directly affects whether or not they can leave work on time.
What GitHub needs to focus on, is doing the hard things, that GitLab and Bitbucket will not be able to do, without serious R&D. I personally think, they should abandon Atom at this point, and use those resources to work on solving harder problems, that Enterprise would gladly pay for. Like better searches, better analytics/reports, better code reviews, and so forth.
As GitHub makes certain parts of the software industry significantly more accessible or cheaper, its complements will succeed commensurately. For example, easy access to, and encouraged proliferation of, skilled talent and robust open source (free) software.
That GitHub provides infrastructure for other startups makes sense, but I'd strongly push back on the idea that VCs will invest in a particular company like GitHub primarily because they think it will improve their investments elsewhere. That's a leap in market forecasting, and the simpler (Occam's Razor) and more rational explanation is that they expect(ed) a good return with some added benefit to "the ecosystem."
I'd be more in favor of this argument if you restated it slightly as, "Investors like investing in companies like GitHub because they can initiate feedback loops with their existing investments that result in mutual prosperity."
In economic terms, it is beneficial for companies to "commoditize their complements." If flights to Miami become cheaper, hotels there can raise their rates. When web browsers became free and ubiquitous, it allowed web applications to prosper on the new influx of people with access to the internet. Google built an empire on this principle.
GitHub commoditizes (or more accurately, pressures costs lower to commoditization for) several complements that would otherwise reduce the profitability and viability of software based businesses. It provides widespread, free access to software as a hosting provider of open source software. It also provides widespread, comparatively low cost access to proven software talent as a sort of professional social network. As the costs associated with one part of an industry fall, the company that causes that decline and the companies that are complementary to those costs benefit greatly.
It is hard to start a new company offering a product for a market inefficiency that has been thoroughly solved for most use cases by commoditized, open source software. However, companies with a complementary relationship to that software will benefit greatly, either because they will have a wider market potential through new customers or because their own costs will drop precipitously. Similarly, developers commoditize themselves and their own costs to a company if they cease differentiating themselves or make it easier to find a supply of them. As an exercise, who do you think is better off in the video game industry - video game studios who make games with a de facto maximum price of $60 (ignoring DLC and in-game purchases), or Microsoft and Sony, who can charge for access to all of those games on an initial (hardware) and monthly (network) basis? Do you think it is in Apple's favor as a company that sells expensive hardware to have developers on their platform charge more or less for their software?
Anyway, my main point was acknowledging that, yes, GitHub makes a lot of other software companies easier through this principle, but no, I don't really agree that venture capitalists have this as an explicit investment strategy. That attributes to them spectacular market forecasting ability and there is a much simpler explanation, which is that they actually believe the company will do well.
Agree. It makes sense to diversify and have interests in other companies that are somewhat related to the wider business.
But... we're talking hundreds of millions of dollars in investment. I believe that's too much cash to be just about subsidizing some other companies.
IMO: GitHub has a business model, a reputation and recurring revenues. They may be a "normal"[1] business, with long term value and stable returns over time.
[1] i.e. not yet another ephemeral app with no plan for monetization that grows 100x then sells for 20B and dies 5 years later).
Sounds a little unlikely tbh. A VC fund exists to invest in a small portfolio of ventures to maximise returns over a fixed period to the Limited Partners who have invested in that particular fund. The Limited Partners aren't donating their money to help the wider ecosystem, and if they suspect the VC firm is investing in businesses without massive profit potential because they might assist the growth of startups in their other VC funds, it's lawyer time.
If that was the case, I think you would expect small investments from VCs and strong calls for sustainable spending. Github had a $100M Series A and a $250M Series B, so the VCs were not playing around when they invested. Even if a VC firm has a $1B fund, a $50M investment is 5% of the fund and has to be a serious investment.
A scrappy low-budget startup can do just fine with a medium sized vm on one of the several places renting vms, running a simple source code control system server. Blowing dozens of megabucks to build a palace in a high-rent district to replace something that's close to free seems, well, dumb.
Github is better. It just is. You don't have to worry about backups, it has great tools built on top of the vcs to make working with it and project management easier. It makes it easier to manage the D part of DVCS, and it lets you easily put your tools into the community.
If you run an open-source project, the community is on Github and that's where they expect to file bugs. Most users/contributors already have their notifications tweaked the way that they like, understand the workflow and website, etc.
At this point, Github is in the same league as Wikipedia in terms of "a fantastic story of technology helping people to create ad-hoc, altruistic collaborations".
The question I'm asking is whether it's better enough to justify a vast investment. New businesses -- startups -- need to take actions to conserve their cash, and sometimes "good enough" is good enough.
It's especially important to get this right when one choice is an unprofitable unicorn-style company with vast cash outflow.
On the back end, software programming tools and Internet-based services make it easy to launch new global software-powered start-ups in many industries—without the need to invest in new infrastructure and train new employees.
That's a huge part of the investment behind Bitcoin's Blockstream. Blockstream pays more more development hours then any other entity by far. There's no direct revenue to be had by improving bitcoin-core.
But without that development we'd be stuck with less security, less features, and a blockchain that took weeks or months to download and verify (instead of days). Bitcoin is only possible because of the massive amount of development going into this unprofitable codebase.
And I believe that Blockstream's investors are well aware that they will never have direct ROI on many of those upgrades.
>The business itself may not be a great business due to the amount of cost it takes to run it -- but it's necessary for the running of other ventures.
The latter half of the statement applies to Stripe as well. The big difference between GitHub and Stripe is that Stripe's business model is more directly tied to the growth of entrepreneurship economy: the more money goes through Stripe-powered businesses, the more money Stripe makes. In GitHub's case, the correlation is less direct and less predictable.
The business itself may not be a great business due to the amount of cost it takes to run it -- but it's necessary for the running of other ventures.
Sort of like highways and non-toll bridges.