Live data from Hacker News

The Cost of Developers

stratechery.com

171–180 of 238 posts

Re: The Cost of Developers

#171

> the App Store dramatically lowered the barriers to entry for developers Is there any platform with higher barriers to entry than the App Store? iOS development requires a proprietary toolchain only available on Apple hardware, programming languages hardly used elsewhere, not to mention the $99/year developer fee and the litany of vague and arbitrarily enforced rules and content guidelines for App Store release.

"Is there any platform with higher barriers to entry than the App Store? "

Just about everything that came before. Getting an app on a phone before meant you had to go through the carrier's store, where they had a 70/30 split, with you getting 30%. There's also selling your software in a box on a store shelf, where the distributor takes 50% off the top.

Re: The Cost of Developers

#172
post #5

Great analysis, although I don't completely agree with this statement: "GitHub, a company that, having raised $350 million in venture capital, was not going to make it as an independent entity." If it is referring to the fact that, in the crazy VC spiral of the startups world, once you have received such a big investment, a sale to a big corp is the only choice, then I sadly agree. But GitHub could have been an indep…

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 ent…

> 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.

Re: The Cost of Developers

#174

Earlier quoted context omitted.

> Seriously? GitHub could have easily gone out for more VC. Making it as an independent company generally involves not taking on more debt on top of a massive pile of debt. What's necessary is, you know, profits and stuff.

1) Equity isn't debt. 2) Profits? You must be kidding. Take a look at some SaaS companies that have gone public in the last 5 years. Few are profitable. It's all about growth of revenue. The profits are way down the line.

Equity is absolutely a form of debt, especially when given to the kinds of VCs that demand high exits.

Maybe if there were VCs who would be happy to own part of the company and be satisfied with the dividends from Github's revenue, it'd be a different story.

Re: The Cost of Developers

#175
post #154

Earlier quoted context omitted.

SAP

SAP isn't "windows desktop software". This whole thread started by saying that Microsoft can leverage its Windows dominance. SAP can either be used as web based software and even thier "Windows" Software is mostly built using thier proprietary cross platform ABAP platform.

I wasn't completely serious about SAP, just meant to gesture to the the general direction of where I see commercial windows software. There are many corporate software vendors I deal with that sell windows software, for an example a suit of Office plugins that connects to SharePoint and do documents versioning.

Another area where I see windows desktop software, and I don't see it goes anywhere soon, is in industrial, medical and labs equipment control.

Re: The Cost of Developers

#176

Earlier quoted context omitted.

If you had a great idea for a non game piece of software, what market would you target? Windows, Web, or mobile? Where are all of the mythical profitable companies making commercial desktop software? Where are all the companies making new internal line of business desktop software?

Much manufacturing and warehouse software is on Windows, with extremely high end costs. ERPs, some are moving to the cloud but so much is still hosted locally. With HUGE annual license fees (often in the millions per year). These systems are slow, unintuitive, and annoying. But they're stable. Companies don't want to risk their existence to change something that sort of works, so you'd be amazed at what a ERP solutio…

But to answer the parent poster's question, if you were developing a new piece of manufacturing/warehouse solution, would you write it as a native windows app or would you target web or mobile? Is the claim that manufacturing/warehouse software is and will be on windows... forever?

Re: The Cost of Developers

#177

Earlier quoted context omitted.

The bigger challenge, if you're a long time Windows user, will be overcoming your own muscle memory. Apple's keyboard layout and shortcuts are similar but different enough to cause problems. It can be quite distracting for instance if you're in the zone and hit the shortcut to replace some text, and then suddenly every window on your desktop disappears!

Biggest problem I have with a mac is the " mark. On a PC (GB keyboard) it's shift-2, on a mac shift-2 is @, and " is shift-'

You can customize your key mappings. I found a dvorak keyboard with esperanto key bindings on the internet. If that exists, I'm sure anything else can be done (assuming you touch type).

Re: The Cost of Developers

#178

MS paid $7.5 Billion for Github, along with its 28 Million users - it's in the first paragraph of the announcement (1) That's around $268 per Github user. Now MS could have probably made a site comparable to Github for 0.1% of that cost - e.g. revive codeplex, pump a few million into the design, backend, git support and user interactions. But it wouldn't have 28M users. That's what they're paying for. That's the hard…

> That's around $268 per Github user.

That seems like a decent deal for MS... Assume on average a user is an active engineer for 5 years (of course this is back-of-the-envelope and not even taking into account the growth of the field, and new engineers are likely to use Github for it's network effects).

Is the value of owning the central platform of an individual engineer worth more than $50 / year? Even if all you did was run basic ads I'd say yes - although I'd expect more services including edging in on Atlassian's turf as an upsell opportunity for MS.

All this, not even counting future value... If Slashdot is still around and somewhat valuable, Github will probably also still be around and somewhat valuable at least a decade from now.

Re: The Cost of Developers

#179
post #156
post #19

Earlier quoted context omitted.

> Sorry, what? Windows still has > 90% market share on desktop. If that isn't "user-based leverage" I don't know what is. At first I thought the same way, but then I started wondering if those users are really meaningful users. Isn't most of the new developer products are either mobile apps or SaaS solutions, leaving aside monsters like adobe? If so, does it really matters that so many users are using windows if they…

> If so, does it really matters that so many users are using windows if they consume apps mainly with their (probably, chrome) browser? Windows is monetizing Telemetry data, which does have meaning and value, even for people that never leave chrome.

As far as we, the general public, know for certain, telemetry is used to gather usability and preferences data, and bugs.

You can avoid using their software if you are concerned about your privacy, that's totally legitimate, You can also warn others about it, if you wish so and believe that it is an important cause. I'll walk most of that walk with you. But I believe you have no justification for your accusation.

Re: The Cost of Developers

#180

MS paid $7.5 Billion for Github, along with its 28 Million users - it's in the first paragraph of the announcement (1) That's around $268 per Github user. Now MS could have probably made a site comparable to Github for 0.1% of that cost - e.g. revive codeplex, pump a few million into the design, backend, git support and user interactions. But it wouldn't have 28M users. That's what they're paying for. That's the hard…

> That's around $268 per Github user. That seems like a decent deal for MS... Assume on average a user is an active engineer for 5 years (of course this is back-of-the-envelope and not even taking into account the growth of the field, and new engineers are likely to use Github for it's network effects). Is the value of owning the central platform of an individual engineer worth more than $50 / year? Even if all you d…

> That seems like a decent deal for MS..

Indeed, as cost of acquisition of an active developer, it doesn't seem either absurdly high or absurdly low.

Post reply on HN