[jillesvangurp]

Correct, github sold out to investors, which then owned them, which then sold them out to get an exit when it turned out the company was in no shape for an IPO which is the only reasonable other exit if you are looking for a 10x ROI on a company that burned through hundreds of millions. Investors were in this for a huge exit and they just got it. In fairness, MS bought a valuable social network of essentially the entire OSS and developer community.

As an independent entity, Github would have had to evolve in a very different way and gone more aggressively after making money, fixing their cost structure, like most of their competitors like Atlassian, Gitlab, and other companies selling developer tools as a SAAS service. So, even though these are comparable companies with comparable offerings, the value proposition is very different. There is no way in hell Gitlab is worth anywhere near what MS just paid for Github.

Being based in silicon valley means Github paid a premium for being there and burned through loads of cash paying for expensive developers, fancy office space, etc. Much cheaper if you are based elsewhere, have less emphasis on wanting to have every OSS project on your platform and more emphasis on selling tools to corporations. This is in a nutshell the strategy for Atlassian and Gitlab. Both also offer freemium layers for OSS but mostly their deal is upselling to their paid offerings. Growth that way is much slower. Though, I would say, Gitlab is now pretty well positioned to succeed where Github struggled.

[jasode]

>, github sold out to investors,

>Investors were in this for a huge exit

I think these excerpts from your comment and also DHH's "VCs need their pound of flesh" are not helpful for readers on how to analyze the situation. They (maybe unintentionally) taint the discussion.

Github is an entity owned by human beings. The founders included Chris Wanstrath, Tom Preston-Werner, PJ Hyett, and Scott Chacon. Therefore, emphasizing that "investors wanted a big exit" is (inadvertently) omitting that those 4 founders may have also wanted a big exit as well. The DHH quote also misdirects people into thinking the VCs are the bad guys. Instead, we have to remember that the 4 founders have to sign off on the documents at closing to get the $350 million. They had to do presentations to convince investors to give them $350 million.

We have to include the founders' thinking about Github's future and not just outsource our frustration to those "evil VCs". VCs cannot give money to a startup if the founders don't want it. (E.g. Craig Newmark refuses VC money for craigslist.org.)

[jillesvangurp]

You are extrapolating an opinion that I don't have. I don't consider the founders, investors, or MS evil. The only mildly controversial thing here is that a big corporation like Microsoft with a long history of hostility towards anything OSS ended up buying the goto gathering place for the OSS community. I don't think that ever was a dream scenario for any of the early founders or early adopters of that company. It will be interesting to see if e.g. Linus Torvalds migrates his linux repository elsewhere.

But obviously, Github was never a charity. It is owned by a group of investors that invested hundreds of millions. An exit in the form of an acquisition or IPO are the only two reasonable happy endings for such a company. This was like this from the day the founders decided to sell ownership to investors. That's what it means to play that game. Usually a first round leads to follow up rounds. Which leads to revenue and ultimately to an IPO. That's the scenario you pitch to investors. In the end it is about shareholder value.

So, normally an acquisition is considered the less ideal outcome of those two. It implies some sort of failure to deliver. Github running without a CEO for a while, reported cash flow issues, and then getting acquired by MS sort of supports that kind of story. I'm assuming the founders did indeed well for themselves. I'm wondering if the same applies to their employee stock option programs since typically those are a bit lower in the pecking order. I don't really know enough about the ownership structure to say anything here but I imagine for most employees ending up working for Microsoft might not have been the dream outcome. And as for most Github users, this probably was a bit of an surprise/shock.

In my case, I'm a paying customer and I see no reason to change that. Love the product so far and MS seem to have the right intentions and strategy to keep it that way.

[jasode]

> I don't consider the founders, investors, or MS evil.

Fair point if I overstated your perspective on VCs.

I don't know if English is your primary language but if you didn't know, the phrase "sold out" is virtually always used in a derogatory manner. When one writes "<X> sold out to <Y>", the <Y> is the proverbial bad guy. It's also parsed as an insult to both <X> and <Y>. E.g. "It's a shame that guitarist sold out to BadMusic Record Company."

It seemed like "sold out to investors" emotionally taints the VCs as bad actors and poisoned the discussion. (Yes, some VCs are bad and unethical.)

In any case, people should think of VCs as one type of financing option. (Bootstrap is another, but that has its own risks and downsides. Bank loans are another finance option but most startup founders have no collateral.) If you don't want to pressure your company into a big exit, do not talk to VCs.

>So, normally an acquisition is considered the less ideal outcome of those two. It implies some sort of failure to deliver.

Maybe. One of the VCs I saw mentioned that acquisitions are actually the more common exits for B2B companies. Github (enterprise) was more B2B than B2C. If this is the consensus view, both a16z and Sequoia may have anticipated acquisition as the most likely (and successful) outcome at the time they invested.

[x0x0]

Well, the founders certainly are (as near as I can tell) letting MSFT and their VCs take heat for a decision the founders alone made.

Taking $350m in VC means you promise to build a stupidly profitable company or sell it to a high bidder. That's the deal, and it's completely understood by anyone in a position to raise nine figures of VC.

[jillesvangurp]

I'm not native indeed but fluent enough to deal with the subtleties. I like to use strong language when making points in public fora. It's a style that is admittedly a bit over dramatic but tends to provoke people a bit more into engaging.

From my dealings with VCs my understanding is that they are mainly incentivized by investing and less by the success of those investments because the timelines are half a decade or more and they reap their rewards long before that happens. Basically the job of a VC is to spend money in some fund. It's a very competitive market to be in. Acquisitions are the tool of choice to make sure under-performing investments are not a total loss. A lot of acquisitions and acquihires are effectively investors consolidating their losses. It's common for them to have stakes in both the acquiring and acquired company. In this case, that's probably not what happened of course. But still, the investors are probably quite happy with this outcome.

[pythonaut_16]

From the standpoint of an end user (which is all of the people complaining about Microsoft acquiring Github) VCs and their money are the bad guys, because the dynamics of their investment is what causes the acquisitions which change the service in ways that I (the end user) do not like.

Not saying they're right or wrong (I think at the very least it's a limited point of view) but that's the argument being made.

[jasode]

>VCs and their money are the bad guys, because the dynamics of their [the VCs] investment

You're repeating the same "outsourcing of blame" to VCs that I was trying to short circuit.

VCs as an abstraction are a convenient target but that doesn't mean they are the correct target.

The founders of Github such as Chris Wanstrath, Tom Preston-Werner, et al have agency and autonomy. Why are they specifically not included in our analysis?

In other words, why aren't we saying this:

>CW & TPW's desire for VC money are the bad guys, because the dynamics of their request for VC investment

It seems wrong that analysis, blame, and frustration seems to gravitate towards VCs and terminates only at the VCs -- instead of the founders who willingly asked the VCs for the money.

Yes, DHH is against VC money. But maybe CW and TPW don't think about Github exactly the same way as DHH thinks about Basecamp.

[ghaff]

VCs are a convenient abstract bogeyman. Yes, VCs do tend to encourage a grow big or go home mentality but it is, as you say, the founders who wanted that sort of investment.

It's also the case that, as soon as someone here suggests that a startup doesn't need to operate in that manner, someone--as certainly as the sun will rise in the east--will post Paul Graham's startup definition and argue that you're not a real startup without a growth-first mentality.

[myWindoonn]

You're super-close. Money is the "bad guy". It might not appear to have agency or autonomy, but that is because humans are not good at perceiving intangible power structures.

In other words, why aren't we saying this:

>Money is the bad guy, because its presence, uneven distribution, cult of worshipers, fungibility, and inherent ties to social power structures are unhealthy influences on the founders of Github

[pm90]

I feel like I should post Francisco D'Anconia's Money speech here....

Money by itself isn't the "bad guy"; its a store of value. Like Technology it can be used for good or bad. We have structured our society to reward greed with unbelievable wealth. Like how can someone spend billions of dollars of their wealth? What is the need for someone to own all that wealth? And yet we all take that as given and aspire to it.

[jchrisa]

Beyond money, it's the logic of commoditization that is the bad guy.

[megaman22]

I can't think of anybody that would be better to acquire GitHub, if it had to be sold off.

At least Microsoft is heavily invested in developer tools, and making them better all the time.

[Retric]

Users love the VC's money. Uber handed out multiple billions of dollars subsidizing both riders and drivers from these massive investments, but even smaller investments often mean the same things. From photo sharing to local delivery most of these services are beloved in large part because they don't need to be directly profitable.

[beauzero]

...yet. "Directly or indirectly profitable" eventually happens.

[rwallace]

Okay, but anyone making that argument should also consider that the existence of VCs and their money may well have been part of the reasoning that led the founders to create the service in the first place. Look at the almost total absence of such useful or interesting services created in places that don't have VCs.

[Clubber]

The term, "github sold out to investors," adequately includes the founders as part of the equation to me. "Sold out," describes all that you mentioned about courting investors and whatnot.

[coldcode]

At where I work we use Github Enterprise, which costs way more than Gitlab Enterprise, is not scalable, and barely cares about enterprise customers like ourselves. Gitlab on the other hand is much cheaper, is highly scalable, runs in AWS, and is responsive to requests. Ultimately to make it as a business you need to provide what the customer wants and is willing to pay. Github never did.

[foobarbazetc]

We switched from GitHub Enterprise to BitBucket Server several years ago and are pretty (80%) happy with the decision. Saved an absolute truck load of money.

The only issue with BitBucket Server is that it (inexplicably) does not have a commit compare view.

https://jira.atlassian.com/plugins/servlet/mobile#issue/BSER...

This is such a massive oversight I have to conclude Atlassian themselves don’t use this product.

It annoys us but not enough to pay GH.

We don’t like GitLab as it was slow when we tested it and the UI wasn’t designed as nicely as the other two (tbh it looks like it was designed by programmers). Something about GitLab feels off and I don’t really know what it is.

If MS changes the cost structure of GH Enterprise we’d probably switch back (haven’t looked into how well it integrates with Jira etc recently).

[gm-conspiracy]

Also, their Sourcetree offering for Windows is quite nice, but the OS X version has big problem with git-flow.

You can tell it is two very different teams working on it.

Atlassian is now like the Cisco of late 90s an early 2000s, an amalgamation of disparate companies' technologies under the umbrella brand.

[mercutio2]

There are many things I find frustrating about BitBucket Server (it’s super slow, for one, it’s insistence on clobbering commit strings during merges is the biggest), but I’ve never wanted a commit compare view; I don’t even know what that is.

Care to explain what you would use it for?

[foobarbazetc]

Just seeing all the commits/changes between two commits.

So with GitHub Enterprise when you push multiple commits, the webhook sends a “compare” URL which is a handy look at everything that got pushed.

It’s not possible to do the same thing with BitBucket Server.

You can compare two tags and get the same view. Or two branches. But not two commits.

[Xaena]

I'd be interested to hear about the slow behavior you're seeing. There are some tips and tricks for improving that.

[mercutio2]

Generally speaking, I have found Atlassian’s tools don’t scale well to megacorp usage. When there are thousands to tens of thousands of users connecting at peak times, stuff just gets extremely slow. Noticeably slower than when we all put up our own team-level Phabricator instances, although I’m obviously not adjusting for Phabricator downtime when inevitably the unofficial team sysadmin went on vacation and the server fell over.

I’m sure people involved in infrastructure are working on making this better, but I don’t have any visibility into that layer.

[Xaena]

From a speed standpoint, I'll say the obvious thing, "it's architecture dependent".

For thousands to tens of thousands of users, most folks need data center with 1+n nodes (w/ larger diminishing returns at 8 nodes). There's a few tips and tricks with that, like pointing CI/CD tools to a smart mirror instead of the primary instance and only mirroring specific build projects. Putting smart mirrors in the same physical location as a primary instance could also help reduce the impact of CPU-intensive requests like git fetch or clone.

Another common issue is apps/plugins. Bitbucket 5.9 added a way to get app diagnostics so infra folks can see if there's a long-running process or other weirdness. Vanilla Bitbucket needs about a 1gb heap (java) and with apps, you want around 4gb (give or take).

If you want a longer read to share, this still has some good info despite being 2~ years old: https://developer.atlassian.com/blog/2016/12/how-we-built-bi...

[u801e]

> The only issue with BitBucket Server is that it (inexplicably) does not have a commit compare view.

Github and Github Enterprise also fail to provide this view. They do provide a compare view for the triple-dot notation (git diff <commit1>...<commit2>) which actually compares the merge-base of commit1 and commit2 with commit2. It's not possible to do the equivalent of git diff commit1..commit2 in the web interface.

[pm90]

> If MS changes the cost structure of GH Enterprise we’d probably switch back (haven’t looked into how well it integrates with Jira etc recently).

Given that they acquired GH mostly for the Open Source parts, I wouldn't bet on that :/. Enterprise was Github's way of making money and that incentive is kinda reduced, if not eliminated entirely.

[Xaena]

Atlassian employee that works with/on Bitbucket here.

We're working on improving that experience. I'd keep a lookout on the next few releases.

[FeloniousHam]

I was on a team evaluating Github and Gitlab for a large corporation (>10000 seats). Github was multiples more expensive than Gitlab, and unwilling to even come close on a counter offer.

I am a fan of Github (and can appreciate their reluctance to deal with Enterprise headaches), but Gitlab was much more engaged and committed to win the contract. Github didn't seem to care whether they got it or not.

[singingfish]

There are quite a few things about gitlab that smell of amateur hour with the way it's put together, but on the whole it's a really good product and I like that enough to make it usable is free open source, and the CI stuff is much nicer than any other I have used (although the way we've put it together, we've hit its limitations a bit).

[2RTZZSro]

You accuse Gitlab of "quite a few [problematic] things" but do not put forward even a single one of those.

[singingfish]

I was on mobile and on my way to bed :P. Lots of page refresh related problems, and likes to exhaust memory from time to time are the two standouts.

[mikekchar]

Have I missed something here? Why are you with Github and not Gitlab?

[blattimwind]

Generally speaking, the set of people who live with the consequences of a decision is not the same set of people who made said decision.

[cease]

The commenter may not be in a position to make the decision to switch. It's also possible that changing would incur high switching costs - although I'm not sure what they'd be.

[coldcode]

We have both, however some teams predate Gitlab option and corporate inertia, politics and a lack of time keep those teams from switching. In a huge company changing anything can be a major pain in the ass.

[wastedhours]

Corporate inertia?

[hsribei]

Gitlab also took VC money and have an exit as their only option. Doubt they'll get as far as GitHub did though, and the option to just grow slowly into a solid independent business is off the table. Probably an acquihire with small return to investors.

[ameister14]

Gitlab is really open about building to a 2020 IPO

"We want to IPO in 2020, specifically on Wednesday November 18. 2020 is five years after the first people got stock options with 4 years of vesting. To IPO we need more than $100m in revenue. To achieve that we want to double Incremental Annual Contract Value (IACV) every year. We focus on an incremental number instead of growth of our Annual Recurring Revenue (ARR) because ARR growth is misleading. So far we achieved the goal of doubling IACV in 2013, 2014, 2015, 2016, and 2017."

[hsribei]

Great find! Here's the source link: https://about.gitlab.com/strategy/#sequence-

[indigochill]

A big difference being Gitlab is open source and can be self-hosted, so you can still run Gitlab on your own terms even after (insert faceless corporate entity here) buys it.

If you care about such things, anyway.

[guscost]

> There is no way in hell Gitlab is worth anywhere near what MS just paid for Github.

On the other hand, Atlassian is worth twice what MS just paid for Github.

[sulam]

> Gitlab is now pretty well positioned to succeed where Github struggled.

Gitlab is also VC funded. In fact it seems like you should be _happy_ that Github is no longer owned by the VCs.